- Email: [email protected]
Prevalence and profile of Salmonella from samples along the production line in Chinese beef processing plants
Food Control 38 (2014) 54e60
Contents lists available at ScienceDirect
Food Control journal homepage: www.elsevier.com/locate/foodcont
Prevalence and profile of Salmonella from samples along the production line in Chinese beef processing plants Pengcheng Dong 1, Lixian Zhu 1, Yanwei Mao, Rongrong Liang, Lebao Niu, Yiming Zhang, Ke Li, Xin Luo* Lab of Beef Processing and Quality Control, College of Food Science and Engineering, Shandong Agricultural University, Tai’an, Shandong 271018, PR China
a r t i c l e i n f o
a b s t r a c t
Article history: Received 18 June 2013 Received in revised form 27 September 2013 Accepted 30 September 2013
This work investigated the prevalence of Salmonella, the serotypes and antibiotic resistance of the isolated strains from four beef processing plants of China. The prevalence of Salmonella in hide (n ¼ 70), feces (n ¼ 70), pre-evisceration carcass (n ¼ 70), post-evisceration carcass (n ¼ 70), post-washing carcass (n ¼ 70), chilled carcass (n ¼ 80), and raw meat (n ¼ 80) samples was 20.0%, 18.6%, 1.4%, 1.4%, 2.9%, 1.3%, and 1.3%, respectively. Among the four plants, there were significant differences in the prevalence of Salmonella on hides and in feces. During the processing, Salmonella was significantly reduced after hide removal. Seven serotypes of Salmonella were identified among the eighty-three isolates. Salmonella Agona was the dominant serotype (p < 0.05, 53.0%), followed by Salmonella Senftenberg (16.9%), Salmonella Meleagridis (10.8%), and Salmonella Derby (9.6%). None of the isolated strains were found to be resistant to sixteen commonly used antimicrobial agents. The results of this study indicate that Salmonella contamination is common in samples along the production line, with S. Agona as dominant serotype. Specific measures should be taken to prevent and/or treat Salmonella contamination in corresponding products in Chinese beef processing plants. Furthermore, the current research might provide baseline information of Salmonella prevalence profile in Chinese beef processing plant, which could be used for the future study. Ó 2013 Elsevier Ltd. All rights reserved.
Keywords: Beef Salmonella Prevalence Serotype Antimicrobial susceptibility
1. Introduction Salmonella is the most common food-borne pathogen in the world (Voetsch et al, 2004), and typhoid and paratyphoid fevers caused by Salmonella are the most common diseases in developing countries and can be fatal under poor sanitary conditions (Rhoades, Duffy, & Koutsoumanis, 2009). Salmonella is the leading pathogen found in China, causing approximately 70%e80% of foodborne diseases, most of which display an association with the consumption of contaminated meat products (Wang, Zheng, & Wang, 2007). The muscle surface of the cattle carcass is sterile, and contamination with Salmonella occurs during in-plant and/or out-plant slaughtering processes (Rivera-Betancourt et al., 2004). The hide is the predominant reservoir of Salmonella, which is mainly transmitted from feces through animaleanimal and/or animalenvironment-animal contact. Because of mechanical tearing, * Corresponding author. Tel./fax: þ86 5388242745. E-mail address: [email protected] (X. Luo). 1 Pengcheng Dong and Lixian Zhu are co-first authors and contributed equally to this work. 0956-7135/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.foodcont.2013.09.066
Salmonella is easily transferred to the carcass during hide removal. Puncture of the bowel and rumen during evisceration can lead to distribution of the pathogen during processing. Environmental and internal cross-contamination may also contribute to the prevalence of foodborne pathogens (Galland, 1997). Because the slaughtering line provides a direct connection between the carcass and the preslaughter animal, it is easily contaminated by processes that distribute pathogens during slaughter. Moreover, without appropriate intervention methods or supervision, the slaughtering line becomes a medium, or even a reservoir, for the direct transmission of pathogens to the carcass, thereby affecting the safety of the resultant products. Several studies in the United States, Northern Ireland, Australia, and Belgium have reported prevalence of Salmonella on cattle carcasses varying between 0.2% and 4.1% (Ghafir et al., 2005; Li, Sherwood, & Logue, 2004; Madden, Espie, Moran, McBride, & Scates, 2001; Vanderlinde, Shay, & Murray, 1998). China has developed the Hazard Analysis Critical Control Points plan to decrease the risk of contamination in processing plants. Adequate microbiological data are needed to assess the effectiveness of control programs. However, the historical mode of farming in China
P. Dong et al. / Food Control 38 (2014) 54e60
in which farmers show great respect for cattle prevented the development of the Chinese beef industry until 1978. Traditionally, the Chinese consume only well-done and not raw or lightly cooked beef. Although there is some published data on the prevalence of Salmonella in retail Chinese foods (Yan et al., 2010; Yang et al., 2013), there is no available information about the contamination profile in products along the processing line of Chinese beef plants. Thus, this study was designed to address the following two aims relevant to beef safety in China: (i) estimation of the incidence of Salmonella in beef processing line, (ii) determination of the serotypes and the antimicrobial susceptibility profiles of the isolated Salmonella strains. 2. Materials and methods 2.1. Experimental design and sample collection A total of 510 samples were collected from four beef processing plants (designated plants A through D) in summer (Julye September, 2010). Three of these plants were located in different cities distributed geographically in the northeast, northwest, and southwest regions of a province in eastern China, and the fourth plant was located in a province in northeast China. All of the commercial plants included in this study were equipped with a production line capable of slaughtering 30e50 animals per hour. Following dressing, the carcasses were washed with cold water before chilling. Six points in the processing line were selected for sampling in this study: (i) hide e samples were collected before the hide removal but after the slaughtering of the animal; (ii) preevisceration carcasses e samples were taken after hide removal; (iii) post-evisceration carcasses e samples were obtained after evisceration; (iv) post-washing carcasses e samples were collected after spraying with water; (v) chilled carcasses e samples were taken from carcasses that had been stored in a chilling room for approximately 24 h; and (vi) meat e samples were collected from cut and deboned carcasses. For the first five points, samples were obtained via gauze swabbing. The sampling sties were chosen according to a previous study, but with a small change in the size of the gauze pads (Breum & Boel, 2010). Briefly, two standard 20 20 cm gauze pads moistened with 0.9% sodium chloride and 0.1% peptone diluents were swabbed at 13 points on the carcass, and the two swabs used for both sides of each carcass were pooled in a stomacher bag. The total area sampled on each side was approximately 2500 cm2. For step (vi), a meat sample approximately of 25 g was randomly collected from beef cuts and placed in an ice storage box. In addition, fecal samples (approximately 10 g) were collected after de-legging. A sterile plastic bag was inverted over the sampler’s hand to prevent contamination of the exposed inner surface of the bag and the bag was inserted inside the rectum to remove fecal material. All samples were transported to the laboratory in coolers containing ice packs. 2.2. Pre-enrichment of the samples In the laboratory, sterile trypticase soy broth (TSB) (Beijing Land Bridge Technology Co., Ltd, Beijing, China) was added to the stomacher bags (BagPageÒ Interscience, France) containing the gauze pads. A total of 100 ml of TSB was added to the swab samples, while the feces (10 g) and meat (25 g) samples were pre-enriched in 90 or 225 ml of TSB, respectively. After being stomached for 1 min at the speed of 8 extrusions per second (BagMixerÒ 400 Interscience, France), all samples were processed following the method described by Barkocy-Gallagher et al. (2002). Briefly, the whole stomacher bags were incubated at 25 C for 3 h and then at 42 C for 6 h. Following incubation, 1 ml of the liquid was pipetted into a
55
1.5 ml tube and stored at 4 C overnight for subsequent immunomagnetic separation. Prior to transfer, an additional enrichment step was performed specifically for the fecal samples, in which 10 ml of the pre-enrichment mixture was transferred to 90 ml of Tetrathionate Broth (TTB), followed by incubation at 37 C for an additional 24 h. After this enrichment, 1 ml of the resulting liquid was pipetted into a 1.5 ml tube and stored as described previously. 2.3. Recovery and confirmation of Salmonella Immunomagnetic separation (IMS) was carried out using immunomagnetic beads coated with an anti-Salmonella antibody (DynabeadsÒ anti-Salmonella, Dynal A.S., Oslo, Norway). The entire IMS procedure was conducted as per the manufacturer’s instructions. After separation, a post-enrichment step was performed for all of the bead-Salmonella complexes: 50 mL of the IMS-bead complex recovered by enrichment was added to 10 ml of RappaporteVassiliadis R10 Broth (RV) (Beijing LandBridge Technology Co., Ltd, Beijing, China), followed by incubation at 37 C for 18 h (Cudjoe & Krona, 1997). After incubation, a full loop of this selectively enriched sample was streaked onto Hektoen enteric (HE) agar (Beijing Land Bridge Technology Co., Ltd.) and CHROMagar Salmonella (CHROMagar Co., Ltd, Paris, France), then incubated for 24 h at 37 C. A minimum of two colonies were isolated and streaked for further purification on MacConkey agar plates (24 h at 37 C) and were confirmed as GB/T 4789.4 (Ministry of health of the Peoples’ Republic of China, 2010) via biochemical and serological methods. The biochemically confirmed isolates were serotyped by KauffmaneWhite classification scheme in Fu Jian Center for Disease Control and Prevention, China. O and H antigens were characterized using slide agglutination with hyperimmune sera (S&A Company, Thailand) and the serotype was assigned following the manufacturer’s instructions. 2.4. Detection of the invA gene The invA gene was detected using a polymerase chain reaction (PCR)-based method described previously (Rahn et al., 1992). The primers (invA Forward: 50 -GTGAAATTATCGCCACGTTCGGGCAA-30 , invA Reverse: 50 -TCATCGCACCGTCAAAGGAACC-30 ) for these assays were commercially synthesized by TaKaRa Biotechnology Co. Ltd (Dalian, China). PCR was carried out in a total volume of 50 mL containing 4 mL of template DNA, 0.2 mL of the forward and reverse primers in total, 0.15 mL of Taq enzyme, 5 mL of PCR buffer, 2.5 mL of MgCl2, 2 mL of dNTPs, and 36.15 mL of nuclease-free water. Positive (Salmonella Typhimurium) and negative control (sterile water) were conducted in the detection procedure. PCR was performed in a DNA thermal cycler. After an initial denaturation step of 7 min at 72 C, 35 cycles of amplification were performed. Each cycle consisted of the following steps: 1 min at 94 C for denaturation, 30 s at 53 C for primer annealing, and 1 min at 72 C for extension, with a final extension at 72 C for 7 min. The samples were analyzed by mixing 5 mL of the reaction mixture with gel loading buffer, followed by resolution via electrophoresis on 2% agarose gels together with the DL1000Ò DNA ladder (TaKaRa Biotechnology Co., Ltd.). Image documentation was carried out with a gel imaging system and viewed on a computer. The PCR products were sequenced by TaKaRa Biotechnology Co. Ltd (Dalian, China). 2.4.1. Antimicrobial susceptibility testing A disk diffusion test was performed according to approved protocols (Clinical and Laboratory Standards Institute, 2006a) to determine the susceptibility of the Salmonella isolates to nalidixic acid, ciprofloxacin, tetracycline, streptomycin, gentamicin, chloramphenicol, amoxicillin/clavulanic acid, ampicillin, cephalotin,
56
P. Dong et al. / Food Control 38 (2014) 54e60
cefoperazone, amikacin, kanamycin, aztreonam, sulfamethoxazole, cefuroxime, and ceftriaxone. The turbidity of the bacterial suspension (in 0.8% NaCl) was adjusted to 0.5 McFarland (1.5 108 CFU/mL). A volume of 100 mL suspension was spread plated onto Mueller Hinton agar (Beijing Land Bridge Technology Co. Ltd.) using a cotton swab. Standard antibiotic disks (Tianhe co., Ltd, Hangzhou, China), each containing a specific concentration of an antibiotic, were then applied. After incubation at 37 C for 18e 24 h, the inhibition zone of each disk was measured in millimeters and the classes of the resistance level were defined as described by the Clinical and Laboratory Standards Institute (2006b). 2.5. Statistical analysis Statistical analysis of the results was performed with PEPI Vernon 11.28 software (Abramson, 2011). To test for differences in prevalence between different processing points and plants, the chi-squared test was used to calculate pair-wise difference in prevalence (a ¼ 0.05). To avoid inflated type I error rates due to multiple comparisons, the pair-wise p values were adjusted using Hommel’s modification of the Bonferroni procedure (Hommel, 1988).
Fig. 1. Occurrence of Salmonella at different sampling points in the four plants. Different letters indicate significant differences among the four plants within the same sampling point (P < 0.05).
3.2. Evaluation of the virulence potential of the isolates via polymerase chain reaction
3. Results 3.1. Salmonella prevalence Of the 510 samples tested, 33 (6.5%) were positive for Salmonella (Table 1). A total of 83 Salmonella isolates were collected from the 33 Salmonella-positive samples. Positive samples were found at all the seven points tested during processing. Among these seven tested, the hides displayed the highest prevalence of Salmonella (20%). The prevalence in the feces, hide, pre-evisceration, post-evisceration, post-washing, chilled carcass, and raw meat samples were 18.6%, 20.0%, 1.4%, 1.4%, 2.9%, 1.3%, and 1.3%, respectively (Table 1). The overall prevalence of Salmonella in the hide and feces samples was significantly greater than in carcasses and raw meat sampled at the same time and was significantly reduced after hide removal (p < 0.05, Table 1). No difference in prevalence was found during the subsequent stages of processing (p > 0.05, Table 1). The overall prevalence among the four plants was significantly different, and most of the positive samples were collected from plant A (22 of 33, Table 1). Surprisingly, no Salmonella was isolated from samples of feces, hides, carcasses and meat in plant C. Significant differences (p < 0.05) were found in the fecal samples, but not in the hide samples between plant A and plant D (Fig. 1). Apart from the hide and feces samples, no significant differences were found in samples from the other stages examined among the four plants (Fig. 1).
The product of gene invA is essential for the bacteria to invade mammalian cells and subsequently lead to foodborne disease. As it is present in most Salmonella and all pathogenic serovars so far (González-Escalona, Brown, & Zhang, 2012), the invA gene is usually used as target gene for Salmonella confirmation. All 83 Salmonella isolates were screened for the invA gene through PCR. The 284 bp DNA fragment was detected from all the 83 reaction mixtures, which confirmed the presence of invA in all isolates (Fig. 2). 3.3. Serotypes of Salmonella isolates Among the 83 Salmonella isolates, a total of seven serotypes were identified, though the serotype could not be determined for three isolates. Table 2 showed the details of the Salmonella serotypes isolated from the carcass, meat, hide, and fecal samples from the four plants. Comparison of the prevalence of the eight serotypes (including the serotype of none-typable) was shown in Fig. 3. S. Agona was the most prevalent serotype isolated from the samples, accounting for 53% of all isolates. The other Salmonella serotypes identified included Senftenberg (16.9%), Meleagidis (10.8%), Derby (9.6%), Kottbus (2.4%), and Calabar (2.4%). Serotype Kingston was isolated relatively rarely (1.2%). Multiple serotypes were recovered from six hide samples and three fecal samples from plants A and D.
Table 1 Prevalence of Salmonella in feces, hide, carcass, and raw meat samples in four beef processing plants. Sampling points
Plant A (%)
Plant B (%)
Plant C (%)
Plant D (%)
Total (%)
Feces Hides Pre-evisceration carcasses Post-evisceration carcasses Post-washing carcasses Chilled carcasses Meat Total
11/20 9/20 0/20 0/20 1/20 1/30 0/30 22/160
0/20 1/20 0/20 1/20 0/20 0/20 1/20 3/140
0/10 0/10 0/10 0/10 0/10 0/10 0/10 0/70
2/20 4/20 1/20 0/20 1/20 0/20 0/20 8/140
13/70 14/70 1/70 1/70 2/70 1/80 1/80 33/510
a, b x, y
(55.0) (45.0) (0.0) (0.0) (5.0) (3.3) (0.0) (13.8)x
(0.0) (5.0) (0.0) (5.0) (0.0) (0.0) (5.0) (2.1)y
Different letters indicate significant differences between sampling points (p < 0.05). Different letters indicate significant differences between plants (p < 0.05).
(0.0) (0.0) (0.0) (0.0) (0.0) (0.0) (0.0) (0.0)y
(10.0) (20.0) (5.0) (0.0) (5.0) (0.0) (0.0) (5.7)xy
(18.6)a (20.0)a (1.4)b (1.4)b (2.9)b (1.3)b (1.3)b (6.5)
P. Dong et al. / Food Control 38 (2014) 54e60
57
Fig. 3. Occurrence of Salmonella in different serotypes. Different letters indicate significant differences between serotypes.
number 42 and 50) presented intermediate resistance to at least one antibiotic (ampicillin-streptomycin-cephalotin and tetracycline-ceftriaxone (Table 4). All 15 isolates showing intermediate resistance were separated from fecal and hide samples (Table 4). Fig. 2. PCR amplification of the invA gene in 38 Salmonella isolates. Number 1e83 represents the 83 Salmonella isolates. Letter “P” represents positive control (Salmonella Typhi). Letter “N” represents negative control and “M” represents molecular weight marker.
3.4. Antimicrobial susceptibility Antibiotic sensitivity testing was performed for the 83 isolates. No resistant isolates were found (Table 3); 15 isolates showed intermediate susceptibility and the remaining 68 isolates were sensitive to all antibiotics. All of the isolates were sensitive to sulfamethoxazole, amikacin, chloramphenicol, cefoperazone, cefuroxime, amoxicillin/clavulanic acid, ciprofloxacin, aztreonam and nalidixic acid. Of the 83 total isolates, 7.2% displayed intermediate resistance to tetracycline and 4.8% to streptomycin; two isolates (2.4%) showed intermediate susceptibility to kanamycin, ampicillin, and cephalotin; and one (1.2%) exhibited intermediate susceptibility to ceftriaxone and gentamicin. Two isolates (strain Table 2 Salmonella serotypes isolated from four beef processing plants (numbers in the brackets indicate the number of isolates with the same serotype). Samples
Plant A (60)
Feces
S. Meleagridis (1) S. Derby (1) S. Agona (9) S. Senftenberg (4) S. Calabar (1) S. Kingston (1) O10/HMC (1) S. Kottbus (2) S. Agona (27) S. Senftenberg (10) S. Calabar (1) O19 H:g,t () (1) O19 H () (1)
Hide
Preevisceration Postevisceration Post-washing S. Agona (1) Chilled S. Agona (1) carcasses Raw meat
Plant B (4)
Plant C Plant D (19) (0) S. Meleagridis (4) S. Derby (1)
S. Meleagridis (1) S. Derby (5) S. Agona (5)
S. Agona (1) S. Derby (1) S. Meleagridis (2)
S. Meleagridis (1)
4. Discussion In-plant harvest practices, which reflect the effectiveness of interventions used in processing plants, can influence the prevalence of Salmonella on carcasses. Feces and hide are considered the main sources of the incoming Salmonella (Rivera-Betancourt et al., 2004), and our methods were specifically designed to estimate the level of contamination in these samples. As expected, the prevalence of Salmonella in feces was much higher than in the other samples (Table 1). Previous studies conducted in Canada, the USA, and the UK demonstrated a prevalence of 2%e6.3% in fecal samples (Abouzeed, Hariharan, Poppe, & Kibenge, 2000; Barkocy-Gallagher et al., 2003; Madden, Murray, & Gilmour, 2007; McEvoy, Doherty, Sheridan, Blair, & McDowell, 2003). Compared to these previous studies, the prevalence of Salmonella in feces from the present study was much higher (Table 1), which might be due to fasting, animal stress, and nutrition management (Abouzeed et al., 2000; Grau, Brownlie, & Smith, 1969). However, for technical limitation, the samples could not be traced back. So the geographical differences of Salmonella prevalence should be further investigated. Table 3 Antimicrobial sensitivity profiles of 83 Salmonella isolates. Antimicrobial agent
Susceptible isolates Resistant
Intermediate
Susceptible
Sulfamethoxazole Kanamycin Amikacin Ampicillin Chloramphenicol Streptomycin Tetracycline Cefoperazone Cefuroxime Cephalotin Ceftriaxone Gentamicin Amoxicillin/clavulanic acid Ciprofloxacin Aztreonam Nalidixic acid
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 2 0 2 0 4 6 0 0 2 1 1 0 0 0 0
83 81 83 81 83 79 77 83 83 81 82 82 83 83 83 83
(2.4%) (2.4%) (4.8%) (7.2%)
(2.4%) (1.2%) (1.2%)
(100%) (97.6%) (100%) (97.6%) (100%) (95.2%) (92.8%) (100%) (100%) (97.6%) (98.8%) (98.8%) (100%) (100%) (100%) (100%)
58
P. Dong et al. / Food Control 38 (2014) 54e60
Table 4 Sources and serotypes of Salmonella isolates with intermediate resistance. Strain number
Sampling point
Serotype
Resistance profile
T17 59 54 40 1 31 36 37 39 41 42 50 68 72 74
Hide Hide Hide Hide Feces Hide Hide Hide Hide Hide Hide Hide Hide Feces Feces
S. Agona S. Agona S. Agona S. Agona S. Meleagridis O19 H:g,t () S. Agona S. Agona S. Agona S. Agona S. Agona S. Agona S. Agona S. Agona S. Agona
Tetracycline Streptomycin Kanamycin Tetracycline Streptomycin Tetracycline Streptomycin Gentamicin Cephalotin Tetracycline AmpicillineStreptomycineCephalotin TetracyclineeCeftriaxone Tetracycline Ampicillin Kanamycin
The prevalence of Salmonella on hide observed in the present study (20.0%) was much lower than we had expected. In a previous study, the rate of isolation of Salmonella-positive samples was low among fecal samples but significantly higher in hide samples (Barkocy-Gallagher et al., 2003). This phenomenon indicates the existence of serious cross-contamination through anima-animal and/or animal-environment-animal contact during large-scale animal processing in the United States (Beach, Murano, & Acuff, 2002). The low prevalence of Salmonella on hide indicated lower crosscontamination levels between individuals in four beef plants. The well-known difference between beef plants in developed country and China is the level of industrialized feeding systems. The smaller herd sizes, geographical isolation of farmers, and lower density of cattle might play a role in preventing cross-contamination between beef individuals in China, which need further investigation. Among the four plants tested, plant A displayed the highest prevalence of Salmonella in hide and fecal samples (p < 0.05), indicating the existence of variation in the Salmonella distribution in different areas of China (Fig. 1). Interestingly, a significant difference (p＜0.05) was observed between fecal samples from plant A and plant D, but not in the hide samples from these plants. This may be a due to varying degrees of cross-transmission in the plants while holding cattle prior to slaughter. Compared to samples of feces collected from the gut, hide samples may be more severely affected by cross-contamination, resulting in the blurring of individual characteristics. It was surprising that no Salmonella was detected in plant C. The reason might be that plant C was located in Inner Mongolia Autonomous Region, which was 1200 km far away from where plant A located. The geographical differences (such as climate and feed) might contribute to the prevalence difference. An additional factor underlying potential differences between these plants was that plant C belonged to a Muslim food products factory. Cows and culled dairy cattle were not slaughtered in this factory. A high prevalence of Salmonella spp. in culled dairy cows at slaughter was found in a previous study (Troutt et al., 2001) and this might explain the relatively low prevalence of Salmonella in plant C. Because of the lack of tracing system, information about the slaughtered cattle (such as breed, source, nutrition and preslaughter management) was hardly profiled. These ratiocinations should be further investigated. A shortage of this study was the less number of samples from plant C. Further studies involving larger sample sizes are necessary to verify the current findings. In-plant practices have an immediate effect on the pathogen populations of the final product. In this study, the detection of Salmonella was significantly reduced following hide removal (P < 0.05), which is consistent with the results of a previous study (Bacon, Sofos, Belk, Hyatt, & Smith, 2002). The hide removal process
plays a two-fold role in pathogen transmission. First, because hide is the main vector of microorganisms, hide removal is also the natural method for the elimination of Salmonella. However, the extreme mechanical force exerted during processing poses a major risk of hide-to-carcass microbial transfer and contamination (Doyle & Erickson, 2006). Hide removal should be considered as a critical step during processing at which additional intervention strategies should be implemented. This process is especially important for Chinese beef slaughterhouses, where showers and other preslaughter intervention methods are rare. The prevalence of Salmonella found in carcasses in the present study was 1.7%, which is slightly higher than the 1.3% reported by Bacon et al. (2002). At in-plant locations, Salmonella is recovered infrequently and can be detected, but not enumerated. Therefore, testing for the prevalence of Salmonella during these processing steps provides minimal useful information for the implementation and maintenance of HACCP systems. In this study, the Salmonella phenotypes were further identified by the detection of invA gene. Because the samples in this research were non-clinical origin, this analysis was also used to find whether invA negative isolates existed in our research. All of the 83 Salmonella isolates with seven different serotypes were positive to the invA gene, which indicated that the negative strains of invA were in a lower frequency. In this study, S. Agona was found to be the most frequent serotype among all of the identified Salmonella isolates, followed by Salmonella Senftenberg, Salmonella Meleagridis, and Salmonella Derby. This result are very different from those of previous studies in which S. Indiana, Salmonella Infantis and Salmonella Enteritidis were shown to be the most common Salmonella serotypes in the chicken industry and markets (Wang et al., 2013; Yang et al., 2013). This disparity can be attributed to the differences in the sampling sources involved in these studies. In China, comparison of the similarities and differences between Salmonella serovars has been difficult due to the lack of systematic epidemiological data on bovine Salmonella infection. Because of its clinical significance, S. Typhimurium has often been isolated in the beef industry (Foley & Lynne, 2008; McEvoy et al., 2003). However, no Salmonella belonging to this serotype were found in our study. This result is consistent with a domestic study demonstrating that S. Typhimurium is seldom found in China (Wang, Ran, Wang, & Li, 2004). Geographical differences may also contribute to these differences. Some studies have shown that the Salmonella virulence plasmid plays an important role in human disease (Guiney et al., 1995). However, the ability to carry virulence plasmids is limited to a few serotypes, such as S. Typhimurium, Salmonella Abortusovis, Salmonella Choleraesuis, Salmonella Dublin, Salmonella Enteritidis, Salmonella Gallinarum-pullorum, and Salmonella Sendai. Fortunately, the Salmonella strains recovered in this study are not on this list of virulent strains. Therefore, with the exception of the S. Derby strain, the serotypes isolated in the present study rarely display the ability to cause human infection. Compared to carcasses and raw meat, there was a greater diversity of serotypes found in fecal and hide samples (Table 2). At plants A and D, the serotypes identified in carcass samples were also found in the respective fecal or hide samples, suggesting that feces and hide were the main source of Salmonella in these processing plants. However, at plant B, the serotypes isolated at different stages during processing were quite different, suggesting that Salmonella contamination may be endogenic at plant B. Among the different plants, the recorded serotypes were quite different from each other, indicating geographical variation between the plants (Table 2). Although serotyping is useful for identifying some differences between strains, more powerful tools, such as molecular typing, are necessary for detailed analysis of these differences
P. Dong et al. / Food Control 38 (2014) 54e60
to build a traceability system among different production processes. The emergence of antimicrobial resistance in foodborne pathogens is a serious public health concern. Drug and multi-drug resistant isolates were found very common in poultry industry (Lu et al., 2011; Wang et al., 2013). No resistant isolate was found in the present study, which indicated low drug resistance of Salmonella in beef plants. The frequency and dosage of antimicrobial agents used in cattle feeding facilities might explain this unique phenomenon, which needs further exploration. According to the USDA, only 25% of small cattle feedlot operations use antimicrobials, and the dosage administered to cattle in feed and water is half that employed in large feedlots, where the prevalence of antimicrobial use is approximately 70% in large-scale cattle feeding facilities (Perez-Montaño et al., 2012). In this study, 15 isolates were found to show intermediate resistance to tetracycline, streptomycin, ampicillin, kanamycin, cephalotin, ceftriaxone, and gentamicin. Among these isolates, 13 were confirmed to be S. Agona (Table 4). Interestingly, there appears to be a correlation between a particular serotype and antimicrobial susceptibility. Resistance to streptomycin and tetracycline has been reported as the most common resistance profile in the beef industry (Dargatz et al., 2003; Perez-Montaño et al., 2012). Although no resistant isolates were detected in the present study, this trend is consistent with our findings given that intermediate resistance to tetracycline (7.2%) and streptomycin (4.8%) was the most frequently observed resistance profile. Considering that tetracycline and streptomycin are widely used in stockbreeding to treat infections and as growth enhancers (Dunlop et al., 1998), more attention should be focused on the supervision and control of antimicrobial use. Furthermore, although antibiotic resistant Salmonella was not isolated in the present study, it is important to monitor the presence of Salmonella during processing. Salmonella was shown to tolerate gastric acidity better when inoculated onto certain foods high in protein and low in fat (Waterman & Small, 1998). Therefore it can cause disease at doses as low as 50 to 100 organisms when consumed as part of a contaminated food source. 5. Conclusions To our knowledge, this is the first study describing the prevalence and characterization of Salmonella in commercial beef processing plants in China. Our results indicate that Salmonella contamination is common in beef processing plants of China, with S. Agona as dominant serotype. Furthermore, the prevalence and profile of Salmonella in beef plant environment (other than beef production line) and the corresponding effects on meat products contamination in beef processing plants needs further investigation. Acknowledgments This research was supported by the National 863 Projects aimed at the control of foodborne pathogenic microorganisms in chilled meat (2012AA10160503), an earmarked fund for the China Agriculture Research System (CARS-38-02A), and the Non-profit Project of Quality Control for Meat Production and Processing (200903012). References Abouzeed, Y. M., Hariharan, H., Poppe, C., & Kibenge, F. S. (2000). Characterization of Salmonella isolates from beef cattle, broiler chickens and human sources on
59
Prince Edward Island. Comparative Immunology Microbiology and Infectious Diseases, 23(4), 253e266. Abramson, J. H. (2011). WINPEPI updated: computer programs for epidemiologists, and their teaching potential. Epidemiologic Perspectives & Innovations: EPþI, 8(1), 1. Bacon, R. T., Sofos, J. N., Belk, K. E., Hyatt, D. R., & Smith, G. C. (2002). Prevalence and antibiotic susceptibility of Salmonella isolated from beef animal hides and carcasses. Journal of Food Protection, 65(2), 284e290. Barkocy-Gallagher, G. A., Arthur, T. M., Rivera-Betancourt, M., Nou, X., Shackelford, S. D., Wheeler, T. L., et al. (2003). Seasonal prevalence of Shiga toxin-producing Escherichia coli, including O157:H7 and non-O157 serotypes, and Salmonella in commercial beef processing plants. Journal of Food Protection, 66(11), 1978e1986. Barkocy-Gallagher, G. A., Berry, E. D., Rivera-Betancourt, M., Arthur, T. M., Nou, X., & Koohmaraie, M. (2002). Development of methods for the recovery of Escherichia coil O157:H7 and Salmonella from beef carcass sponge samples and bovine fecal and hide samples. Journal of Food Protection, 65(10), 1527e1534. Beach, J. C., Murano, E. A., & Acuff, G. R. (2002). Prevalence of Salmonella and Campylobacter in beef cattle from transport to slaughter. Journal of Food Protection, 65(11), 1687e1693. Breum, S.Ø., & Boel, J. (2010). Prevalence of Escherichia coli O157 and verocytotoxin producing E. coli (VTEC) on Danish beef carcasses. International Journal of Food Microbiology, 141(1e2), 90e96. Clinical and Laboratory Standards Institute. (2006a). Performance standards for antimicrobial disk susceptibility tests. Clinical and Laboratory Standards Institute document M2eA9 (Approved standard 9th ed.). 940 West Valley Road, Suite 1400, Wayne, Pennsylvania 19087-1898 USA: Clinical and Laboratory Standards Institute, ISBN 1-56238-586-0. Clinical and Laboratory Standards Institute. (2006b). Performance standards for antimicrobial susceptibility testing. Sixteenth Informational Supplement. CLSI document M100eS16. 940 West Valley Road, Suite 1400, Wayne, Pennsylvania 19087-1898 USA: Clinical and Laboratory Standards Institute. ISBN 1-56238556-7. Cudjoe, K. S., & Krona, R. (1997). Detection of Salmonella from raw food samples using Dynabeads anti-Salmonella and a conventional reference method. International Journal of Food Microbiology, 37(1), 55e62. Dargatz, D. A., Fedorka-Cray, P. J., Ladely, S. R., Kopral, C. A., Ferris, K. E., & Headrick, M. L. (2003). Prevalence and antimicrobial susceptibility of Salmonella spp. isolates from US cattle in feedlots in 1999 and 2000. Journal of Applied Microbiology, 95(4), 753e761. Doyle, M. P., & Erickson, M. C. (2006). Reducing the carriage of foodborne pathogens in livestock and poultry. Poultry Science, 85(6), 960e973. Dunlop, R. H., McEwen, S. A., Meek, A. H., Black, W. D., Clarke, R. C., & Friendship, R. M. (1998). Individual and group antimicrobial usage rates on 34 farrow-to-finish swine farms in Ontario, Canada. Preventive Veterinary Medicine, 34(4), 247e264. Foley, S. L., & Lynne, A. M. (2008). Food animal-associated Salmonella challenges: pathogenicity and antimicrobial resistance. Journal of Animal Science, 86(14 Suppl.), E173eE187. Galland, J. C. (1997). Risks and prevention of contamination of beef carcasses during the slaughter process in the United States of America. Revue Scientifique et Technique (International Office of Epizootics), 16(2), 395e404. Ghafir, Y., China, B., Korsak, N., Dierick, K., Collard, J.-M., Godard, C., et al. (2005). Belgian surveillance plans to assess changes in Salmonella prevalence in meat at different production stages. Journal of Food Protection, 68(11), 2269e2277. González-Escalona, N., Brown, E. W., & Zhang, G. (2012). Development and evaluation of a multiplex real-time PCR (qPCR) assay targeting ttrRSBCA locus and invA gene for accurate detection of Salmonella spp. in fresh produce and eggs. Food Research International, 48(1), 202e208. Grau, F. H., Brownlie, L. E., & Smith, M. G. (1969). Effects of food intake on numbers of Salmonellae and Escherichia coli in rumen and faeces of sheep. The Journal of Applied Bacteriology, 32(1), 112e117. Guiney, D. G., Fang, F. C., Krause, M., Libby, S., Buchmeier, N. A., & Fierer, J. (1995). Biology and clinical significance of virulence plasmids in Salmonella serovars. Clinical Infectious Diseases, 21(Suppl. 2), S146eS151. Hommel, G. (1988). A stage wise rejective multiple test procedure based on a modified Bonferroni test. Physical Sciences Biometrika, 75, 383e386. Li, Q., Sherwood, J. S., & Logue, C. M. (2004). The prevalence of Listeria, Salmonella, Escherichia coli and E. coli O157:H7 on bison carcasses during processing. Food Microbiology, 21(6), 791e799. Lu, Y., Wu, C.-M., Wu, G.-J., Zhao, H.-Y., He, T., Cao, X.-Y., et al. (2011). Prevalence of antimicrobial resistance among Salmonella isolates from chicken in China. Foodborne Pathogens and Disease, 8(1), 45e53. Madden, R. H., Espie, W. E., Moran, L., McBride, J., & Scates, P. (2001). Occurrence of Escherichia coli O157:H7, Listeria monocytogenes, Salmonella and Campylobacter spp. on beef carcasses in Northern Ireland. Meat Science, 58(4), 343e346. Madden, R. H., Murray, K. A., & Gilmour, A. (2007). Carriage of four bacterial pathogens by beef cattle in Northern Ireland at time of slaughter. Letters in Applied Microbiology, 44(2), 115e119. McEvoy, J. M., Doherty, A. M., Sheridan, J. J., Blair, I. S., & McDowell, D. A. (2003). The prevalence of Salmonella spp. in bovine faecal, rumen and carcass samples at a commercial abattoir. Journal of Applied Microbiology, 94(4), 693e700. Ministry of health of the Peoples’ Republic of China. (2010). National food safety standard food microbiological examination: Salmonella. GB 4789.4-2010. South Xizhimenwai Road, Xicheng District, Beijing, China: Ministry of Hygiene.
60
P. Dong et al. / Food Control 38 (2014) 54e60
Perez-Montaño, J. A., Gonzalez-Aguilar, D., Barba, J., Pacheco-Gallardo, C., CamposBravo, C. A., Garcia, S., et al. (2012). Frequency and antimicrobial resistance of Salmonella serotypes on beef carcasses at small abattoirs in Jalisco State, Mexico. Journal of Food Protection, 75(5), 867e873. Rahn, K., De Grandis, S. A., Clarke, R. C., McEwen, S. A., Galán, J. E., Ginocchio, C., et al. (1992). Amplification of an invA gene sequence of Salmonella typhimurium by polymerase chain reaction as a specific method of detection of Salmonella. Molecular and Cellular Probes, 6(4), 271e279. Rhoades, J. R., Duffy, G., & Koutsoumanis, K. (2009). Prevalence and concentration of verocytotoxigenic Escherichia coli, Salmonella enterica and Listeria monocytogenes in the beef production chain: a review. Food Microbiology, 26(4), 357e376. Rivera-Betancourt, M., Shackelford, S. D., Arthur, T. M., Westmoreland, K. E., Bellinger, G., Rossman, M., et al. (2004). Prevalence of Escherichia coli O157:H7, Listeria monocytogenes, and Salmonella in two geographically distant commercial beef processing plants in the United States. Journal of Food Protection, 67(2), 295e302. Troutt, H. F., Galland, J. C., Osburn, B. I., Brewer, R. L., Braun, R. K., Schmitz, J. A., et al. (2001). Prevalence of Salmonella spp in cull (market) dairy cows at slaughter. Journal of the American Veterinary Medical Association, 219(9), 1212e1215. Vanderlinde, P. B., Shay, B., & Murray, J. (1998). Microbiological quality of Australian beef carcass meat and frozen bulk packed beef. Journal of Food Protection, 61(4), 437e443.
Voetsch, A. C., Van Gilder, T. J., Angulo, F. J., Farley, M. M., Shallow, S., Marcus, R., et al. (2004). FoodNet estimate of the burden of illness caused by non typhoidal Salmonella infections in the United States. Clinical Infectious Diseases, 38(Suppl. 3), S127eS134. Wang, M., Ran, L., Wang, Z., & Li, Z. (2004). Study on national active monitoring for food borne pathogens and antimicrobial resistance in China 2001. Journal of Hygiene Research, 33(1), 49e54. Wang, H., Ye, K., Wei, X., Cao, J., Xu, X., & Zhou, G. (2013). Occurrence, antimicrobial resistance and biofilm formation of Salmonella isolates from a chicken slaughter plant in China. Food Control, 33(2), 378e384. Wang, L., Zheng, R., & Wang, J. (2007). Risk assessment of Salmonella in foods of animal production. Chinese Journal of Animal Quarantine, 24, 23e25. Waterman, S. R., & Small, P. L. (1998). Acid-sensitive enteric pathogens are protected from killing under extremely acidic conditions of pH 2.5 when they are inoculated onto certain solid food sources. Applied and Environmental Microbiology, 64(10), 3882e3886. Yang, B., Qiao, L., Zhang, X., Cui, Y., Xia, X., Cui, S., et al. (2013). Serotyping, antimicrobial susceptibility, pulse field gel electrophoresis analysis of Salmonella isolates from retail foods in Henan Province, China. Food Control, 32(1), 228e235. Yan, H., Li, L., Alam, M. J., Shinoda, S., Miyoshi, S., & Shi, L. (2010). Prevalence and antimicrobial resistance of Salmonella in retail foods in northern China. International Journal of Food Microbiology, 143(3), 230e234.
Contents lists available at ScienceDirect
Food Control journal homepage: www.elsevier.com/locate/foodcont
Prevalence and profile of Salmonella from samples along the production line in Chinese beef processing plants Pengcheng Dong 1, Lixian Zhu 1, Yanwei Mao, Rongrong Liang, Lebao Niu, Yiming Zhang, Ke Li, Xin Luo* Lab of Beef Processing and Quality Control, College of Food Science and Engineering, Shandong Agricultural University, Tai’an, Shandong 271018, PR China
a r t i c l e i n f o
a b s t r a c t
Article history: Received 18 June 2013 Received in revised form 27 September 2013 Accepted 30 September 2013
This work investigated the prevalence of Salmonella, the serotypes and antibiotic resistance of the isolated strains from four beef processing plants of China. The prevalence of Salmonella in hide (n ¼ 70), feces (n ¼ 70), pre-evisceration carcass (n ¼ 70), post-evisceration carcass (n ¼ 70), post-washing carcass (n ¼ 70), chilled carcass (n ¼ 80), and raw meat (n ¼ 80) samples was 20.0%, 18.6%, 1.4%, 1.4%, 2.9%, 1.3%, and 1.3%, respectively. Among the four plants, there were significant differences in the prevalence of Salmonella on hides and in feces. During the processing, Salmonella was significantly reduced after hide removal. Seven serotypes of Salmonella were identified among the eighty-three isolates. Salmonella Agona was the dominant serotype (p < 0.05, 53.0%), followed by Salmonella Senftenberg (16.9%), Salmonella Meleagridis (10.8%), and Salmonella Derby (9.6%). None of the isolated strains were found to be resistant to sixteen commonly used antimicrobial agents. The results of this study indicate that Salmonella contamination is common in samples along the production line, with S. Agona as dominant serotype. Specific measures should be taken to prevent and/or treat Salmonella contamination in corresponding products in Chinese beef processing plants. Furthermore, the current research might provide baseline information of Salmonella prevalence profile in Chinese beef processing plant, which could be used for the future study. Ó 2013 Elsevier Ltd. All rights reserved.
Keywords: Beef Salmonella Prevalence Serotype Antimicrobial susceptibility
1. Introduction Salmonella is the most common food-borne pathogen in the world (Voetsch et al, 2004), and typhoid and paratyphoid fevers caused by Salmonella are the most common diseases in developing countries and can be fatal under poor sanitary conditions (Rhoades, Duffy, & Koutsoumanis, 2009). Salmonella is the leading pathogen found in China, causing approximately 70%e80% of foodborne diseases, most of which display an association with the consumption of contaminated meat products (Wang, Zheng, & Wang, 2007). The muscle surface of the cattle carcass is sterile, and contamination with Salmonella occurs during in-plant and/or out-plant slaughtering processes (Rivera-Betancourt et al., 2004). The hide is the predominant reservoir of Salmonella, which is mainly transmitted from feces through animaleanimal and/or animalenvironment-animal contact. Because of mechanical tearing, * Corresponding author. Tel./fax: þ86 5388242745. E-mail address: [email protected] (X. Luo). 1 Pengcheng Dong and Lixian Zhu are co-first authors and contributed equally to this work. 0956-7135/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.foodcont.2013.09.066
Salmonella is easily transferred to the carcass during hide removal. Puncture of the bowel and rumen during evisceration can lead to distribution of the pathogen during processing. Environmental and internal cross-contamination may also contribute to the prevalence of foodborne pathogens (Galland, 1997). Because the slaughtering line provides a direct connection between the carcass and the preslaughter animal, it is easily contaminated by processes that distribute pathogens during slaughter. Moreover, without appropriate intervention methods or supervision, the slaughtering line becomes a medium, or even a reservoir, for the direct transmission of pathogens to the carcass, thereby affecting the safety of the resultant products. Several studies in the United States, Northern Ireland, Australia, and Belgium have reported prevalence of Salmonella on cattle carcasses varying between 0.2% and 4.1% (Ghafir et al., 2005; Li, Sherwood, & Logue, 2004; Madden, Espie, Moran, McBride, & Scates, 2001; Vanderlinde, Shay, & Murray, 1998). China has developed the Hazard Analysis Critical Control Points plan to decrease the risk of contamination in processing plants. Adequate microbiological data are needed to assess the effectiveness of control programs. However, the historical mode of farming in China
P. Dong et al. / Food Control 38 (2014) 54e60
in which farmers show great respect for cattle prevented the development of the Chinese beef industry until 1978. Traditionally, the Chinese consume only well-done and not raw or lightly cooked beef. Although there is some published data on the prevalence of Salmonella in retail Chinese foods (Yan et al., 2010; Yang et al., 2013), there is no available information about the contamination profile in products along the processing line of Chinese beef plants. Thus, this study was designed to address the following two aims relevant to beef safety in China: (i) estimation of the incidence of Salmonella in beef processing line, (ii) determination of the serotypes and the antimicrobial susceptibility profiles of the isolated Salmonella strains. 2. Materials and methods 2.1. Experimental design and sample collection A total of 510 samples were collected from four beef processing plants (designated plants A through D) in summer (Julye September, 2010). Three of these plants were located in different cities distributed geographically in the northeast, northwest, and southwest regions of a province in eastern China, and the fourth plant was located in a province in northeast China. All of the commercial plants included in this study were equipped with a production line capable of slaughtering 30e50 animals per hour. Following dressing, the carcasses were washed with cold water before chilling. Six points in the processing line were selected for sampling in this study: (i) hide e samples were collected before the hide removal but after the slaughtering of the animal; (ii) preevisceration carcasses e samples were taken after hide removal; (iii) post-evisceration carcasses e samples were obtained after evisceration; (iv) post-washing carcasses e samples were collected after spraying with water; (v) chilled carcasses e samples were taken from carcasses that had been stored in a chilling room for approximately 24 h; and (vi) meat e samples were collected from cut and deboned carcasses. For the first five points, samples were obtained via gauze swabbing. The sampling sties were chosen according to a previous study, but with a small change in the size of the gauze pads (Breum & Boel, 2010). Briefly, two standard 20 20 cm gauze pads moistened with 0.9% sodium chloride and 0.1% peptone diluents were swabbed at 13 points on the carcass, and the two swabs used for both sides of each carcass were pooled in a stomacher bag. The total area sampled on each side was approximately 2500 cm2. For step (vi), a meat sample approximately of 25 g was randomly collected from beef cuts and placed in an ice storage box. In addition, fecal samples (approximately 10 g) were collected after de-legging. A sterile plastic bag was inverted over the sampler’s hand to prevent contamination of the exposed inner surface of the bag and the bag was inserted inside the rectum to remove fecal material. All samples were transported to the laboratory in coolers containing ice packs. 2.2. Pre-enrichment of the samples In the laboratory, sterile trypticase soy broth (TSB) (Beijing Land Bridge Technology Co., Ltd, Beijing, China) was added to the stomacher bags (BagPageÒ Interscience, France) containing the gauze pads. A total of 100 ml of TSB was added to the swab samples, while the feces (10 g) and meat (25 g) samples were pre-enriched in 90 or 225 ml of TSB, respectively. After being stomached for 1 min at the speed of 8 extrusions per second (BagMixerÒ 400 Interscience, France), all samples were processed following the method described by Barkocy-Gallagher et al. (2002). Briefly, the whole stomacher bags were incubated at 25 C for 3 h and then at 42 C for 6 h. Following incubation, 1 ml of the liquid was pipetted into a
55
1.5 ml tube and stored at 4 C overnight for subsequent immunomagnetic separation. Prior to transfer, an additional enrichment step was performed specifically for the fecal samples, in which 10 ml of the pre-enrichment mixture was transferred to 90 ml of Tetrathionate Broth (TTB), followed by incubation at 37 C for an additional 24 h. After this enrichment, 1 ml of the resulting liquid was pipetted into a 1.5 ml tube and stored as described previously. 2.3. Recovery and confirmation of Salmonella Immunomagnetic separation (IMS) was carried out using immunomagnetic beads coated with an anti-Salmonella antibody (DynabeadsÒ anti-Salmonella, Dynal A.S., Oslo, Norway). The entire IMS procedure was conducted as per the manufacturer’s instructions. After separation, a post-enrichment step was performed for all of the bead-Salmonella complexes: 50 mL of the IMS-bead complex recovered by enrichment was added to 10 ml of RappaporteVassiliadis R10 Broth (RV) (Beijing LandBridge Technology Co., Ltd, Beijing, China), followed by incubation at 37 C for 18 h (Cudjoe & Krona, 1997). After incubation, a full loop of this selectively enriched sample was streaked onto Hektoen enteric (HE) agar (Beijing Land Bridge Technology Co., Ltd.) and CHROMagar Salmonella (CHROMagar Co., Ltd, Paris, France), then incubated for 24 h at 37 C. A minimum of two colonies were isolated and streaked for further purification on MacConkey agar plates (24 h at 37 C) and were confirmed as GB/T 4789.4 (Ministry of health of the Peoples’ Republic of China, 2010) via biochemical and serological methods. The biochemically confirmed isolates were serotyped by KauffmaneWhite classification scheme in Fu Jian Center for Disease Control and Prevention, China. O and H antigens were characterized using slide agglutination with hyperimmune sera (S&A Company, Thailand) and the serotype was assigned following the manufacturer’s instructions. 2.4. Detection of the invA gene The invA gene was detected using a polymerase chain reaction (PCR)-based method described previously (Rahn et al., 1992). The primers (invA Forward: 50 -GTGAAATTATCGCCACGTTCGGGCAA-30 , invA Reverse: 50 -TCATCGCACCGTCAAAGGAACC-30 ) for these assays were commercially synthesized by TaKaRa Biotechnology Co. Ltd (Dalian, China). PCR was carried out in a total volume of 50 mL containing 4 mL of template DNA, 0.2 mL of the forward and reverse primers in total, 0.15 mL of Taq enzyme, 5 mL of PCR buffer, 2.5 mL of MgCl2, 2 mL of dNTPs, and 36.15 mL of nuclease-free water. Positive (Salmonella Typhimurium) and negative control (sterile water) were conducted in the detection procedure. PCR was performed in a DNA thermal cycler. After an initial denaturation step of 7 min at 72 C, 35 cycles of amplification were performed. Each cycle consisted of the following steps: 1 min at 94 C for denaturation, 30 s at 53 C for primer annealing, and 1 min at 72 C for extension, with a final extension at 72 C for 7 min. The samples were analyzed by mixing 5 mL of the reaction mixture with gel loading buffer, followed by resolution via electrophoresis on 2% agarose gels together with the DL1000Ò DNA ladder (TaKaRa Biotechnology Co., Ltd.). Image documentation was carried out with a gel imaging system and viewed on a computer. The PCR products were sequenced by TaKaRa Biotechnology Co. Ltd (Dalian, China). 2.4.1. Antimicrobial susceptibility testing A disk diffusion test was performed according to approved protocols (Clinical and Laboratory Standards Institute, 2006a) to determine the susceptibility of the Salmonella isolates to nalidixic acid, ciprofloxacin, tetracycline, streptomycin, gentamicin, chloramphenicol, amoxicillin/clavulanic acid, ampicillin, cephalotin,
56
P. Dong et al. / Food Control 38 (2014) 54e60
cefoperazone, amikacin, kanamycin, aztreonam, sulfamethoxazole, cefuroxime, and ceftriaxone. The turbidity of the bacterial suspension (in 0.8% NaCl) was adjusted to 0.5 McFarland (1.5 108 CFU/mL). A volume of 100 mL suspension was spread plated onto Mueller Hinton agar (Beijing Land Bridge Technology Co. Ltd.) using a cotton swab. Standard antibiotic disks (Tianhe co., Ltd, Hangzhou, China), each containing a specific concentration of an antibiotic, were then applied. After incubation at 37 C for 18e 24 h, the inhibition zone of each disk was measured in millimeters and the classes of the resistance level were defined as described by the Clinical and Laboratory Standards Institute (2006b). 2.5. Statistical analysis Statistical analysis of the results was performed with PEPI Vernon 11.28 software (Abramson, 2011). To test for differences in prevalence between different processing points and plants, the chi-squared test was used to calculate pair-wise difference in prevalence (a ¼ 0.05). To avoid inflated type I error rates due to multiple comparisons, the pair-wise p values were adjusted using Hommel’s modification of the Bonferroni procedure (Hommel, 1988).
Fig. 1. Occurrence of Salmonella at different sampling points in the four plants. Different letters indicate significant differences among the four plants within the same sampling point (P < 0.05).
3.2. Evaluation of the virulence potential of the isolates via polymerase chain reaction
3. Results 3.1. Salmonella prevalence Of the 510 samples tested, 33 (6.5%) were positive for Salmonella (Table 1). A total of 83 Salmonella isolates were collected from the 33 Salmonella-positive samples. Positive samples were found at all the seven points tested during processing. Among these seven tested, the hides displayed the highest prevalence of Salmonella (20%). The prevalence in the feces, hide, pre-evisceration, post-evisceration, post-washing, chilled carcass, and raw meat samples were 18.6%, 20.0%, 1.4%, 1.4%, 2.9%, 1.3%, and 1.3%, respectively (Table 1). The overall prevalence of Salmonella in the hide and feces samples was significantly greater than in carcasses and raw meat sampled at the same time and was significantly reduced after hide removal (p < 0.05, Table 1). No difference in prevalence was found during the subsequent stages of processing (p > 0.05, Table 1). The overall prevalence among the four plants was significantly different, and most of the positive samples were collected from plant A (22 of 33, Table 1). Surprisingly, no Salmonella was isolated from samples of feces, hides, carcasses and meat in plant C. Significant differences (p < 0.05) were found in the fecal samples, but not in the hide samples between plant A and plant D (Fig. 1). Apart from the hide and feces samples, no significant differences were found in samples from the other stages examined among the four plants (Fig. 1).
The product of gene invA is essential for the bacteria to invade mammalian cells and subsequently lead to foodborne disease. As it is present in most Salmonella and all pathogenic serovars so far (González-Escalona, Brown, & Zhang, 2012), the invA gene is usually used as target gene for Salmonella confirmation. All 83 Salmonella isolates were screened for the invA gene through PCR. The 284 bp DNA fragment was detected from all the 83 reaction mixtures, which confirmed the presence of invA in all isolates (Fig. 2). 3.3. Serotypes of Salmonella isolates Among the 83 Salmonella isolates, a total of seven serotypes were identified, though the serotype could not be determined for three isolates. Table 2 showed the details of the Salmonella serotypes isolated from the carcass, meat, hide, and fecal samples from the four plants. Comparison of the prevalence of the eight serotypes (including the serotype of none-typable) was shown in Fig. 3. S. Agona was the most prevalent serotype isolated from the samples, accounting for 53% of all isolates. The other Salmonella serotypes identified included Senftenberg (16.9%), Meleagidis (10.8%), Derby (9.6%), Kottbus (2.4%), and Calabar (2.4%). Serotype Kingston was isolated relatively rarely (1.2%). Multiple serotypes were recovered from six hide samples and three fecal samples from plants A and D.
Table 1 Prevalence of Salmonella in feces, hide, carcass, and raw meat samples in four beef processing plants. Sampling points
Plant A (%)
Plant B (%)
Plant C (%)
Plant D (%)
Total (%)
Feces Hides Pre-evisceration carcasses Post-evisceration carcasses Post-washing carcasses Chilled carcasses Meat Total
11/20 9/20 0/20 0/20 1/20 1/30 0/30 22/160
0/20 1/20 0/20 1/20 0/20 0/20 1/20 3/140
0/10 0/10 0/10 0/10 0/10 0/10 0/10 0/70
2/20 4/20 1/20 0/20 1/20 0/20 0/20 8/140
13/70 14/70 1/70 1/70 2/70 1/80 1/80 33/510
a, b x, y
(55.0) (45.0) (0.0) (0.0) (5.0) (3.3) (0.0) (13.8)x
(0.0) (5.0) (0.0) (5.0) (0.0) (0.0) (5.0) (2.1)y
Different letters indicate significant differences between sampling points (p < 0.05). Different letters indicate significant differences between plants (p < 0.05).
(0.0) (0.0) (0.0) (0.0) (0.0) (0.0) (0.0) (0.0)y
(10.0) (20.0) (5.0) (0.0) (5.0) (0.0) (0.0) (5.7)xy
(18.6)a (20.0)a (1.4)b (1.4)b (2.9)b (1.3)b (1.3)b (6.5)
P. Dong et al. / Food Control 38 (2014) 54e60
57
Fig. 3. Occurrence of Salmonella in different serotypes. Different letters indicate significant differences between serotypes.
number 42 and 50) presented intermediate resistance to at least one antibiotic (ampicillin-streptomycin-cephalotin and tetracycline-ceftriaxone (Table 4). All 15 isolates showing intermediate resistance were separated from fecal and hide samples (Table 4). Fig. 2. PCR amplification of the invA gene in 38 Salmonella isolates. Number 1e83 represents the 83 Salmonella isolates. Letter “P” represents positive control (Salmonella Typhi). Letter “N” represents negative control and “M” represents molecular weight marker.
3.4. Antimicrobial susceptibility Antibiotic sensitivity testing was performed for the 83 isolates. No resistant isolates were found (Table 3); 15 isolates showed intermediate susceptibility and the remaining 68 isolates were sensitive to all antibiotics. All of the isolates were sensitive to sulfamethoxazole, amikacin, chloramphenicol, cefoperazone, cefuroxime, amoxicillin/clavulanic acid, ciprofloxacin, aztreonam and nalidixic acid. Of the 83 total isolates, 7.2% displayed intermediate resistance to tetracycline and 4.8% to streptomycin; two isolates (2.4%) showed intermediate susceptibility to kanamycin, ampicillin, and cephalotin; and one (1.2%) exhibited intermediate susceptibility to ceftriaxone and gentamicin. Two isolates (strain Table 2 Salmonella serotypes isolated from four beef processing plants (numbers in the brackets indicate the number of isolates with the same serotype). Samples
Plant A (60)
Feces
S. Meleagridis (1) S. Derby (1) S. Agona (9) S. Senftenberg (4) S. Calabar (1) S. Kingston (1) O10/HMC (1) S. Kottbus (2) S. Agona (27) S. Senftenberg (10) S. Calabar (1) O19 H:g,t () (1) O19 H () (1)
Hide
Preevisceration Postevisceration Post-washing S. Agona (1) Chilled S. Agona (1) carcasses Raw meat
Plant B (4)
Plant C Plant D (19) (0) S. Meleagridis (4) S. Derby (1)
S. Meleagridis (1) S. Derby (5) S. Agona (5)
S. Agona (1) S. Derby (1) S. Meleagridis (2)
S. Meleagridis (1)
4. Discussion In-plant harvest practices, which reflect the effectiveness of interventions used in processing plants, can influence the prevalence of Salmonella on carcasses. Feces and hide are considered the main sources of the incoming Salmonella (Rivera-Betancourt et al., 2004), and our methods were specifically designed to estimate the level of contamination in these samples. As expected, the prevalence of Salmonella in feces was much higher than in the other samples (Table 1). Previous studies conducted in Canada, the USA, and the UK demonstrated a prevalence of 2%e6.3% in fecal samples (Abouzeed, Hariharan, Poppe, & Kibenge, 2000; Barkocy-Gallagher et al., 2003; Madden, Murray, & Gilmour, 2007; McEvoy, Doherty, Sheridan, Blair, & McDowell, 2003). Compared to these previous studies, the prevalence of Salmonella in feces from the present study was much higher (Table 1), which might be due to fasting, animal stress, and nutrition management (Abouzeed et al., 2000; Grau, Brownlie, & Smith, 1969). However, for technical limitation, the samples could not be traced back. So the geographical differences of Salmonella prevalence should be further investigated. Table 3 Antimicrobial sensitivity profiles of 83 Salmonella isolates. Antimicrobial agent
Susceptible isolates Resistant
Intermediate
Susceptible
Sulfamethoxazole Kanamycin Amikacin Ampicillin Chloramphenicol Streptomycin Tetracycline Cefoperazone Cefuroxime Cephalotin Ceftriaxone Gentamicin Amoxicillin/clavulanic acid Ciprofloxacin Aztreonam Nalidixic acid
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 2 0 2 0 4 6 0 0 2 1 1 0 0 0 0
83 81 83 81 83 79 77 83 83 81 82 82 83 83 83 83
(2.4%) (2.4%) (4.8%) (7.2%)
(2.4%) (1.2%) (1.2%)
(100%) (97.6%) (100%) (97.6%) (100%) (95.2%) (92.8%) (100%) (100%) (97.6%) (98.8%) (98.8%) (100%) (100%) (100%) (100%)
58
P. Dong et al. / Food Control 38 (2014) 54e60
Table 4 Sources and serotypes of Salmonella isolates with intermediate resistance. Strain number
Sampling point
Serotype
Resistance profile
T17 59 54 40 1 31 36 37 39 41 42 50 68 72 74
Hide Hide Hide Hide Feces Hide Hide Hide Hide Hide Hide Hide Hide Feces Feces
S. Agona S. Agona S. Agona S. Agona S. Meleagridis O19 H:g,t () S. Agona S. Agona S. Agona S. Agona S. Agona S. Agona S. Agona S. Agona S. Agona
Tetracycline Streptomycin Kanamycin Tetracycline Streptomycin Tetracycline Streptomycin Gentamicin Cephalotin Tetracycline AmpicillineStreptomycineCephalotin TetracyclineeCeftriaxone Tetracycline Ampicillin Kanamycin
The prevalence of Salmonella on hide observed in the present study (20.0%) was much lower than we had expected. In a previous study, the rate of isolation of Salmonella-positive samples was low among fecal samples but significantly higher in hide samples (Barkocy-Gallagher et al., 2003). This phenomenon indicates the existence of serious cross-contamination through anima-animal and/or animal-environment-animal contact during large-scale animal processing in the United States (Beach, Murano, & Acuff, 2002). The low prevalence of Salmonella on hide indicated lower crosscontamination levels between individuals in four beef plants. The well-known difference between beef plants in developed country and China is the level of industrialized feeding systems. The smaller herd sizes, geographical isolation of farmers, and lower density of cattle might play a role in preventing cross-contamination between beef individuals in China, which need further investigation. Among the four plants tested, plant A displayed the highest prevalence of Salmonella in hide and fecal samples (p < 0.05), indicating the existence of variation in the Salmonella distribution in different areas of China (Fig. 1). Interestingly, a significant difference (p＜0.05) was observed between fecal samples from plant A and plant D, but not in the hide samples from these plants. This may be a due to varying degrees of cross-transmission in the plants while holding cattle prior to slaughter. Compared to samples of feces collected from the gut, hide samples may be more severely affected by cross-contamination, resulting in the blurring of individual characteristics. It was surprising that no Salmonella was detected in plant C. The reason might be that plant C was located in Inner Mongolia Autonomous Region, which was 1200 km far away from where plant A located. The geographical differences (such as climate and feed) might contribute to the prevalence difference. An additional factor underlying potential differences between these plants was that plant C belonged to a Muslim food products factory. Cows and culled dairy cattle were not slaughtered in this factory. A high prevalence of Salmonella spp. in culled dairy cows at slaughter was found in a previous study (Troutt et al., 2001) and this might explain the relatively low prevalence of Salmonella in plant C. Because of the lack of tracing system, information about the slaughtered cattle (such as breed, source, nutrition and preslaughter management) was hardly profiled. These ratiocinations should be further investigated. A shortage of this study was the less number of samples from plant C. Further studies involving larger sample sizes are necessary to verify the current findings. In-plant practices have an immediate effect on the pathogen populations of the final product. In this study, the detection of Salmonella was significantly reduced following hide removal (P < 0.05), which is consistent with the results of a previous study (Bacon, Sofos, Belk, Hyatt, & Smith, 2002). The hide removal process
plays a two-fold role in pathogen transmission. First, because hide is the main vector of microorganisms, hide removal is also the natural method for the elimination of Salmonella. However, the extreme mechanical force exerted during processing poses a major risk of hide-to-carcass microbial transfer and contamination (Doyle & Erickson, 2006). Hide removal should be considered as a critical step during processing at which additional intervention strategies should be implemented. This process is especially important for Chinese beef slaughterhouses, where showers and other preslaughter intervention methods are rare. The prevalence of Salmonella found in carcasses in the present study was 1.7%, which is slightly higher than the 1.3% reported by Bacon et al. (2002). At in-plant locations, Salmonella is recovered infrequently and can be detected, but not enumerated. Therefore, testing for the prevalence of Salmonella during these processing steps provides minimal useful information for the implementation and maintenance of HACCP systems. In this study, the Salmonella phenotypes were further identified by the detection of invA gene. Because the samples in this research were non-clinical origin, this analysis was also used to find whether invA negative isolates existed in our research. All of the 83 Salmonella isolates with seven different serotypes were positive to the invA gene, which indicated that the negative strains of invA were in a lower frequency. In this study, S. Agona was found to be the most frequent serotype among all of the identified Salmonella isolates, followed by Salmonella Senftenberg, Salmonella Meleagridis, and Salmonella Derby. This result are very different from those of previous studies in which S. Indiana, Salmonella Infantis and Salmonella Enteritidis were shown to be the most common Salmonella serotypes in the chicken industry and markets (Wang et al., 2013; Yang et al., 2013). This disparity can be attributed to the differences in the sampling sources involved in these studies. In China, comparison of the similarities and differences between Salmonella serovars has been difficult due to the lack of systematic epidemiological data on bovine Salmonella infection. Because of its clinical significance, S. Typhimurium has often been isolated in the beef industry (Foley & Lynne, 2008; McEvoy et al., 2003). However, no Salmonella belonging to this serotype were found in our study. This result is consistent with a domestic study demonstrating that S. Typhimurium is seldom found in China (Wang, Ran, Wang, & Li, 2004). Geographical differences may also contribute to these differences. Some studies have shown that the Salmonella virulence plasmid plays an important role in human disease (Guiney et al., 1995). However, the ability to carry virulence plasmids is limited to a few serotypes, such as S. Typhimurium, Salmonella Abortusovis, Salmonella Choleraesuis, Salmonella Dublin, Salmonella Enteritidis, Salmonella Gallinarum-pullorum, and Salmonella Sendai. Fortunately, the Salmonella strains recovered in this study are not on this list of virulent strains. Therefore, with the exception of the S. Derby strain, the serotypes isolated in the present study rarely display the ability to cause human infection. Compared to carcasses and raw meat, there was a greater diversity of serotypes found in fecal and hide samples (Table 2). At plants A and D, the serotypes identified in carcass samples were also found in the respective fecal or hide samples, suggesting that feces and hide were the main source of Salmonella in these processing plants. However, at plant B, the serotypes isolated at different stages during processing were quite different, suggesting that Salmonella contamination may be endogenic at plant B. Among the different plants, the recorded serotypes were quite different from each other, indicating geographical variation between the plants (Table 2). Although serotyping is useful for identifying some differences between strains, more powerful tools, such as molecular typing, are necessary for detailed analysis of these differences
P. Dong et al. / Food Control 38 (2014) 54e60
to build a traceability system among different production processes. The emergence of antimicrobial resistance in foodborne pathogens is a serious public health concern. Drug and multi-drug resistant isolates were found very common in poultry industry (Lu et al., 2011; Wang et al., 2013). No resistant isolate was found in the present study, which indicated low drug resistance of Salmonella in beef plants. The frequency and dosage of antimicrobial agents used in cattle feeding facilities might explain this unique phenomenon, which needs further exploration. According to the USDA, only 25% of small cattle feedlot operations use antimicrobials, and the dosage administered to cattle in feed and water is half that employed in large feedlots, where the prevalence of antimicrobial use is approximately 70% in large-scale cattle feeding facilities (Perez-Montaño et al., 2012). In this study, 15 isolates were found to show intermediate resistance to tetracycline, streptomycin, ampicillin, kanamycin, cephalotin, ceftriaxone, and gentamicin. Among these isolates, 13 were confirmed to be S. Agona (Table 4). Interestingly, there appears to be a correlation between a particular serotype and antimicrobial susceptibility. Resistance to streptomycin and tetracycline has been reported as the most common resistance profile in the beef industry (Dargatz et al., 2003; Perez-Montaño et al., 2012). Although no resistant isolates were detected in the present study, this trend is consistent with our findings given that intermediate resistance to tetracycline (7.2%) and streptomycin (4.8%) was the most frequently observed resistance profile. Considering that tetracycline and streptomycin are widely used in stockbreeding to treat infections and as growth enhancers (Dunlop et al., 1998), more attention should be focused on the supervision and control of antimicrobial use. Furthermore, although antibiotic resistant Salmonella was not isolated in the present study, it is important to monitor the presence of Salmonella during processing. Salmonella was shown to tolerate gastric acidity better when inoculated onto certain foods high in protein and low in fat (Waterman & Small, 1998). Therefore it can cause disease at doses as low as 50 to 100 organisms when consumed as part of a contaminated food source. 5. Conclusions To our knowledge, this is the first study describing the prevalence and characterization of Salmonella in commercial beef processing plants in China. Our results indicate that Salmonella contamination is common in beef processing plants of China, with S. Agona as dominant serotype. Furthermore, the prevalence and profile of Salmonella in beef plant environment (other than beef production line) and the corresponding effects on meat products contamination in beef processing plants needs further investigation. Acknowledgments This research was supported by the National 863 Projects aimed at the control of foodborne pathogenic microorganisms in chilled meat (2012AA10160503), an earmarked fund for the China Agriculture Research System (CARS-38-02A), and the Non-profit Project of Quality Control for Meat Production and Processing (200903012). References Abouzeed, Y. M., Hariharan, H., Poppe, C., & Kibenge, F. S. (2000). Characterization of Salmonella isolates from beef cattle, broiler chickens and human sources on
59
Prince Edward Island. Comparative Immunology Microbiology and Infectious Diseases, 23(4), 253e266. Abramson, J. H. (2011). WINPEPI updated: computer programs for epidemiologists, and their teaching potential. Epidemiologic Perspectives & Innovations: EPþI, 8(1), 1. Bacon, R. T., Sofos, J. N., Belk, K. E., Hyatt, D. R., & Smith, G. C. (2002). Prevalence and antibiotic susceptibility of Salmonella isolated from beef animal hides and carcasses. Journal of Food Protection, 65(2), 284e290. Barkocy-Gallagher, G. A., Arthur, T. M., Rivera-Betancourt, M., Nou, X., Shackelford, S. D., Wheeler, T. L., et al. (2003). Seasonal prevalence of Shiga toxin-producing Escherichia coli, including O157:H7 and non-O157 serotypes, and Salmonella in commercial beef processing plants. Journal of Food Protection, 66(11), 1978e1986. Barkocy-Gallagher, G. A., Berry, E. D., Rivera-Betancourt, M., Arthur, T. M., Nou, X., & Koohmaraie, M. (2002). Development of methods for the recovery of Escherichia coil O157:H7 and Salmonella from beef carcass sponge samples and bovine fecal and hide samples. Journal of Food Protection, 65(10), 1527e1534. Beach, J. C., Murano, E. A., & Acuff, G. R. (2002). Prevalence of Salmonella and Campylobacter in beef cattle from transport to slaughter. Journal of Food Protection, 65(11), 1687e1693. Breum, S.Ø., & Boel, J. (2010). Prevalence of Escherichia coli O157 and verocytotoxin producing E. coli (VTEC) on Danish beef carcasses. International Journal of Food Microbiology, 141(1e2), 90e96. Clinical and Laboratory Standards Institute. (2006a). Performance standards for antimicrobial disk susceptibility tests. Clinical and Laboratory Standards Institute document M2eA9 (Approved standard 9th ed.). 940 West Valley Road, Suite 1400, Wayne, Pennsylvania 19087-1898 USA: Clinical and Laboratory Standards Institute, ISBN 1-56238-586-0. Clinical and Laboratory Standards Institute. (2006b). Performance standards for antimicrobial susceptibility testing. Sixteenth Informational Supplement. CLSI document M100eS16. 940 West Valley Road, Suite 1400, Wayne, Pennsylvania 19087-1898 USA: Clinical and Laboratory Standards Institute. ISBN 1-56238556-7. Cudjoe, K. S., & Krona, R. (1997). Detection of Salmonella from raw food samples using Dynabeads anti-Salmonella and a conventional reference method. International Journal of Food Microbiology, 37(1), 55e62. Dargatz, D. A., Fedorka-Cray, P. J., Ladely, S. R., Kopral, C. A., Ferris, K. E., & Headrick, M. L. (2003). Prevalence and antimicrobial susceptibility of Salmonella spp. isolates from US cattle in feedlots in 1999 and 2000. Journal of Applied Microbiology, 95(4), 753e761. Doyle, M. P., & Erickson, M. C. (2006). Reducing the carriage of foodborne pathogens in livestock and poultry. Poultry Science, 85(6), 960e973. Dunlop, R. H., McEwen, S. A., Meek, A. H., Black, W. D., Clarke, R. C., & Friendship, R. M. (1998). Individual and group antimicrobial usage rates on 34 farrow-to-finish swine farms in Ontario, Canada. Preventive Veterinary Medicine, 34(4), 247e264. Foley, S. L., & Lynne, A. M. (2008). Food animal-associated Salmonella challenges: pathogenicity and antimicrobial resistance. Journal of Animal Science, 86(14 Suppl.), E173eE187. Galland, J. C. (1997). Risks and prevention of contamination of beef carcasses during the slaughter process in the United States of America. Revue Scientifique et Technique (International Office of Epizootics), 16(2), 395e404. Ghafir, Y., China, B., Korsak, N., Dierick, K., Collard, J.-M., Godard, C., et al. (2005). Belgian surveillance plans to assess changes in Salmonella prevalence in meat at different production stages. Journal of Food Protection, 68(11), 2269e2277. González-Escalona, N., Brown, E. W., & Zhang, G. (2012). Development and evaluation of a multiplex real-time PCR (qPCR) assay targeting ttrRSBCA locus and invA gene for accurate detection of Salmonella spp. in fresh produce and eggs. Food Research International, 48(1), 202e208. Grau, F. H., Brownlie, L. E., & Smith, M. G. (1969). Effects of food intake on numbers of Salmonellae and Escherichia coli in rumen and faeces of sheep. The Journal of Applied Bacteriology, 32(1), 112e117. Guiney, D. G., Fang, F. C., Krause, M., Libby, S., Buchmeier, N. A., & Fierer, J. (1995). Biology and clinical significance of virulence plasmids in Salmonella serovars. Clinical Infectious Diseases, 21(Suppl. 2), S146eS151. Hommel, G. (1988). A stage wise rejective multiple test procedure based on a modified Bonferroni test. Physical Sciences Biometrika, 75, 383e386. Li, Q., Sherwood, J. S., & Logue, C. M. (2004). The prevalence of Listeria, Salmonella, Escherichia coli and E. coli O157:H7 on bison carcasses during processing. Food Microbiology, 21(6), 791e799. Lu, Y., Wu, C.-M., Wu, G.-J., Zhao, H.-Y., He, T., Cao, X.-Y., et al. (2011). Prevalence of antimicrobial resistance among Salmonella isolates from chicken in China. Foodborne Pathogens and Disease, 8(1), 45e53. Madden, R. H., Espie, W. E., Moran, L., McBride, J., & Scates, P. (2001). Occurrence of Escherichia coli O157:H7, Listeria monocytogenes, Salmonella and Campylobacter spp. on beef carcasses in Northern Ireland. Meat Science, 58(4), 343e346. Madden, R. H., Murray, K. A., & Gilmour, A. (2007). Carriage of four bacterial pathogens by beef cattle in Northern Ireland at time of slaughter. Letters in Applied Microbiology, 44(2), 115e119. McEvoy, J. M., Doherty, A. M., Sheridan, J. J., Blair, I. S., & McDowell, D. A. (2003). The prevalence of Salmonella spp. in bovine faecal, rumen and carcass samples at a commercial abattoir. Journal of Applied Microbiology, 94(4), 693e700. Ministry of health of the Peoples’ Republic of China. (2010). National food safety standard food microbiological examination: Salmonella. GB 4789.4-2010. South Xizhimenwai Road, Xicheng District, Beijing, China: Ministry of Hygiene.
60
P. Dong et al. / Food Control 38 (2014) 54e60
Perez-Montaño, J. A., Gonzalez-Aguilar, D., Barba, J., Pacheco-Gallardo, C., CamposBravo, C. A., Garcia, S., et al. (2012). Frequency and antimicrobial resistance of Salmonella serotypes on beef carcasses at small abattoirs in Jalisco State, Mexico. Journal of Food Protection, 75(5), 867e873. Rahn, K., De Grandis, S. A., Clarke, R. C., McEwen, S. A., Galán, J. E., Ginocchio, C., et al. (1992). Amplification of an invA gene sequence of Salmonella typhimurium by polymerase chain reaction as a specific method of detection of Salmonella. Molecular and Cellular Probes, 6(4), 271e279. Rhoades, J. R., Duffy, G., & Koutsoumanis, K. (2009). Prevalence and concentration of verocytotoxigenic Escherichia coli, Salmonella enterica and Listeria monocytogenes in the beef production chain: a review. Food Microbiology, 26(4), 357e376. Rivera-Betancourt, M., Shackelford, S. D., Arthur, T. M., Westmoreland, K. E., Bellinger, G., Rossman, M., et al. (2004). Prevalence of Escherichia coli O157:H7, Listeria monocytogenes, and Salmonella in two geographically distant commercial beef processing plants in the United States. Journal of Food Protection, 67(2), 295e302. Troutt, H. F., Galland, J. C., Osburn, B. I., Brewer, R. L., Braun, R. K., Schmitz, J. A., et al. (2001). Prevalence of Salmonella spp in cull (market) dairy cows at slaughter. Journal of the American Veterinary Medical Association, 219(9), 1212e1215. Vanderlinde, P. B., Shay, B., & Murray, J. (1998). Microbiological quality of Australian beef carcass meat and frozen bulk packed beef. Journal of Food Protection, 61(4), 437e443.
Voetsch, A. C., Van Gilder, T. J., Angulo, F. J., Farley, M. M., Shallow, S., Marcus, R., et al. (2004). FoodNet estimate of the burden of illness caused by non typhoidal Salmonella infections in the United States. Clinical Infectious Diseases, 38(Suppl. 3), S127eS134. Wang, M., Ran, L., Wang, Z., & Li, Z. (2004). Study on national active monitoring for food borne pathogens and antimicrobial resistance in China 2001. Journal of Hygiene Research, 33(1), 49e54. Wang, H., Ye, K., Wei, X., Cao, J., Xu, X., & Zhou, G. (2013). Occurrence, antimicrobial resistance and biofilm formation of Salmonella isolates from a chicken slaughter plant in China. Food Control, 33(2), 378e384. Wang, L., Zheng, R., & Wang, J. (2007). Risk assessment of Salmonella in foods of animal production. Chinese Journal of Animal Quarantine, 24, 23e25. Waterman, S. R., & Small, P. L. (1998). Acid-sensitive enteric pathogens are protected from killing under extremely acidic conditions of pH 2.5 when they are inoculated onto certain solid food sources. Applied and Environmental Microbiology, 64(10), 3882e3886. Yang, B., Qiao, L., Zhang, X., Cui, Y., Xia, X., Cui, S., et al. (2013). Serotyping, antimicrobial susceptibility, pulse field gel electrophoresis analysis of Salmonella isolates from retail foods in Henan Province, China. Food Control, 32(1), 228e235. Yan, H., Li, L., Alam, M. J., Shinoda, S., Miyoshi, S., & Shi, L. (2010). Prevalence and antimicrobial resistance of Salmonella in retail foods in northern China. International Journal of Food Microbiology, 143(3), 230e234.