Using molecular techniques for rapid detection of Salmonella serovars in frozen chicken and chicken products collected from Riyadh, Saudi Arabia

The current study was aimed to investigate the incidence of different Salmonella serovars in chicken products either from local or imported source. A total of 152 samples of chicken and chicken products were collected from different retail establishment markets in Riyadh, KSA including 38 local whole frozen chickens, 62 imported whole frozen chickens, 22 whole poultry eggs and 30 local chicken cuts samples and examined by standard microbiological techniques (SMT). Salmonella isolation revealed a total percentage of 5.92%; chicken cuts revealed a high incidence among the examined samples (10%), followed by local frozen chickens and imported frozen chicken samples with incidence of 7.89 and 4.83%, respectively. For this experiment, the whole chicken eggs were negative for Salmonella species by SMT. Salmonella enteritidis was dominating among the recovered Salmonella serovars, followed by Salmonella typhimurium, while only two strains of Salmonella agona and Salmonella newport were isolated. The PCR assay combined with RappaportVassiliadis (RV) selective broth (PCR-RV) for the detection of Salmonella species in the collected field samples revealed the same positive samples directly from the imported frozen chickens and whole chicken eggs which gave negative results by SMT. Thus PCR-RV technique is rapid, time saving and applicable to detect Salmonella serovars directly from chicken samples.


INTRODUCTION
Salmonella species live in the intestinal tracts of warm and cold blooded animals. Some species are ubiquitous; other species are specifically adapted to a particular host. It is the major causes of food-borne disease throughout the world (Altekruse et al., 1999;Humphrey, 2002;Schlundt, 2002;Wang et al., 2008). Salmonella infected chickens represent a source of pathogens for humans, causing severe illness and even death. It is estimated that 16 million new cases of typhoid fever occur each year around the world, mostly in developing country (D'Aoust, 1994;Parry et al., 2002;Dimitrov et al., 2007); the infection is characterized by a variety of clinical manifestations ranging from high-grade fever to complications including "encephalopathy, peritonitis, perforation and hemorrhage". The commonest serotypes causing disease in humans are Salmonella enteritidis and Salmonella typhimurium (Baggesen et al., 2002;Aktas et al., 2007). Multi-resistant S. typhimurium definitive phage type (DT) 104 strains are responsible for a high number of infections in humans and are primarily zoonotic in origin (Gatto et al., 2006). Today, it is widely spread and is considered pandemic. Egg associated Salmonellosis is an important public health problem in the United States and several European countries. S. enteritidis silently infects the ovaries of healthy appearing hens and contaminates the egg before the shells are formed and if the eggs are eaten raw or undercooked, the bacterium can cause illness.
Imported birds and animals may serve to introduce different Salmonella species to the local area that can cause new and devastating outbreaks (Altekruse et al., 1999;Sareyyupoglu et al., 2007;Yu et al., 2008). The isolation and identification of salmonellae from clinical samples by traditional cultural techniques requires laborious procedures which can last up to 7 days (Stone et al., 1994), so there is a need for the development of innovative methods for the rapid identification of Salmonella food-borne pathogen to over come these drawbacks. Molecular techniques such as Polymerase Chain Reaction (PCR) especially by using selective broth culture have been invaluable tools for the detection of different Salmonella species (Oliveira et al., 2003). When multiple target genes need to be amplified, multiplex PCR (MPCR) can be performed and may provide a simple and sensitive tool for the simultaneous detection of multiple pathogenic bacteria (Soumet et al., 1999).
Investigation of S. enteritidis and Salmonella typhimurium among Salmonella enterica serovars in chickens and chicken products collected from Riyadh, King Saudi Arabia (KSA) using conventional and molecular techniques (PCR using selective broth culture) was the major strategy of this study. Moussa et al. 613

Bacteria and reagents
The bacterial reference strains used in this study were illustrated in (Table 1). The materials, chemicals and reagents used in this study were purchased from Sigma Chemicals Co. (St. Louis, MO. USA) unless otherwise specified. PCR reagents were purchased from Promega (Madison, WI, USA).

Samples collection
During the summer of 2009, a total of 152 samples of poultry and poultry products were collected from different retail establishment markets in Riyadh, KSA including, 38 local whole frozen chickens, 62 imported whole frozen chickens, 22 whole poultry eggs and 30 local chicken cut samples (Liver, framed and Giblets). We also used 10 negative control field samples (5 frozen chicken and 5 poultry eggs) collected from young birds a few hours after hatching, and these birds coming from breading flocks were continuously monitored for salmonella by standard microbiological techniques. All samples were transported to the laboratory under refrigerated conditions where they were processed and bacteriologically examined immediately.

Standard microbiological techniques for Salmonella detection and identification (SMT)
The standard microbiological techniques for detection of different Salmonella serovars conducted according to ISO 6579 (2002); 25 g of poultry composite samples were homogenized in a stomacher (Bag Mixer 400, Interscience, France) for 1 to 2 min in 225 ml of buffered peptone water (BPW) and then incubated under aerobic conditions at 37°C for 16 -20 h followed by selective enrichment of 0.1 in 10 ml of Rappaport -Vassiliadis (RV) broth. The RV broth was incubated at 42°C for 18-24 h. The broth was then sub cultured onto Xylose Lysine Desoxycholate agar (XLD) agar, Hektoen Enteric agar and Salmonella Chromogenic Agar then incubated at 37°C for 18 -24 h. All media were supplied by Oxoid, Basingstoke, UK. Presumptive positive colonies (non lactose fermentative with suitable colony morphology) were identified morphologically, biochemically, serologically by slide agglutination test using polyvalent and monovalent somatic (O), virulence (Vi) and tube agglutination test for flageller (H) antigens (Difco Laboratories, Detroit, Michigan, USA) and Enzyme Linked Immuno-Sorbant Assay (ELISA) (Reveal Salmonella test kits systems, Neogen Corporation). 1 ml of BPW which had been incubated at 37°C was saved for the PCR-Non Selective test (PCR-NS) and 1 ml of the 37°C RV broth for the PCR-RV test.

Extraction of DNA
The standard and bacteriologically positive strains were grown in 10 ml Tryptic Soya Broth (TSB) at 37°C for 24 h. The overnight cultures were centrifuged at 3000 rpm for 5 min and the supernatant were decanted carefully. The bacterial pellets were washed three times with phosphate buffer saline pH 7.2 and resuspended in 400 µL tris-EDTA buffer (pH 8.0) and heated in water bath at 100°C for 20 min. There were left to cool at room temperature and centrifuged at 14,000 rpm for 10 min. An aliquot of 5 µL of the supernatant was used as template DNA in the PCR. While the extraction of DNA from the field samples enriched in RV broth was carried out by the same method reported by Oliveira et al. (2003). Table 2. Oligonucleotide primers sequences used for amplification of DNA for the detection of Salmonella species according to (Oliveira et al., 2002).

Oligonucleotide primers
Four sets of primer pairs were used; the first was 139 -141, specific for the invA gene located on the salmonella pathogenicity island 1 which is highly conserved in Salmonella species and encodes a type III secretion system that exports proteins in response to bacterial contact with epithelial cells. (Galan et al., 1992;Rahn et al., 1992); the second was Fli15-Typ04, specific for the fliC gene found in S. typhimurium (Soumet et al., 1999), the third was A058-A01, specific for the sefA gene found in S. enteritidis (Doran et al., 1996). The fourth was SdiA1 and SdiA2 specific to genus Salmonella species according to Konstantia et al. (2006), Primer sequences and specificity, G+C content, annealing temperature and size of the amplified product were summarized in the Tables 2 and 3.

DNA amplification
Four independent reactions, each with one set of primers, were made for each DNA template. PCR amplifications were performed in a final volume of 50 µL in micro-amplification tubes (PCR tubes). The reaction mixtures consisted of 5 µL of the DNA template, 5 µL 10x PCR buffer (75 mM Tris-HCl, pH 9.0, 2 mM MgCl2, 50 mM KCl, 20 mM (NH4)2SO4), 1 µL dNTPs (40 µM), 1 µL (1U Ampli Taq DNA polymerase), 1 µL (25 pmol) from the forward and reverse primers of both primer pairs and the volume of the reaction mixture was completed to 50 µL using DDW. The thermal cycler was adjusted as follows: Initial denaturation at 94°C for 5 min, followed by 35 cycles of (denaturation at 94°C for 1 min, annealing as shown in Table 3 for 1 min and extension at 72°C for 1 min). Final extension was carried out at 72°C for 10 min and the PCR products were stored in the thermal cycler at 4°C until they were collected.

Agarose gel electrophoresis
The PCR products were tested for positive amplification by agarose gel electrophoresis previously reported by Sambrook et al. (1989) using suitable molecular weight markers.

Standard microbiological techniques
The standard microbiological techniques revealed positive isolation of 9 Salmonella serovars (5.92%) out of 152 examined poultry samples and 3 of them (7.89%) were isolated from 38 local frozen chickens and identified biochemically and serologically as S. enteritidis. Also, 3 Salmonella serovars (4.83%) were isolated from 62 imported frozen chickens and identified as Salmonella agona, Salmonella newport and S. enteritidis. While the other 3 Salmonella serovars (10%) were isolated from 30 local chicken cuts and identified by SMT as S. enteritidis (one strain) and S. typhimurium (two strains). The whole chicken eggs were negative for isolation of Salmonella serovars by SMT (Table 4).

Molecular typing using PCR
The specificity of the oligonucleotid primers as well as typing of the recovered Salmonella serovars from SMT  were carried out by testing of all the recovered Salmonella strains in addition to the standard positive and standard negative strains with PCR, using four types of primer pairs targeting for (invA, sdiA, sefA and fliC genes). The specificity of such genes was mentioned in Tables 2 and 3. Whereas all Salmonella serovars were positive for amplification of 284 and 274 bp fragments of invA and sdiA genes, all non Salmonella serovars were negative (Figures 1, 2a and 2b). Only S. enteritidis gave positive amplification of 484 bp fragments of sefA gene; on the other hand, all S. typhimurium were positive for amplification of 620 bp fragments of fliC gene while all Salmonella serovars were negative for the presences of sefA and fliC genes (Figures 3a and b). All the examined field samples with SMT as well as the negative control field samples were tested by PCR using the same primer pairs after selective enrichment on RV broth. All bacteriologically positive samples (100%) were positive by PCR and amplification of 284 and 274 bp fragments specific for invA and sdiA genes were observed. In addition, 2 samples (1.32%) previously identified as negative samples with SMT were positive with PCR using the two primer pairs (Tables 5 and 6). All S. enteritidis positive samples with SMT were positive for amplification of 484 bp fragments specific for sefA gene, moreover, PCR could detect S. enteritidis in 2 field samples (one imported frozen chickens and one whole chicken egg) previously identified as negative for Salmonella species with SMT (Tables 5 and 6). On the other hand, all S. typhimurium positive samples with SMT were positive for amplification of 620 bp fragments specific for fliC gene found in S. typhimurium (Tables 5 and 6). The negative control field samples were negative for the PCR assay and no amplification could be detected with the four primer pairs.

DISCUSSION
Poultry are one of the most important reservoirs of Salmonellae that can be transmitted to humans through the food-chain. The commonest serotypes causing disease in humans are S. enteritidis and S. typhimurium (Aktas et al., 2007). The detection of Salmonella species by regulatory agencies is still primarily based on standard microbiological techniques which may take up to 7 days to confirm the results (Stone et al., 1994). The earlier a foodborne outbreak is suspected, the faster the source of the pathogen can be identified, and the sooner the public can regain confidence in the food supply (Bhagwat and Lauer, 2004). The current study was aimed to investigate the incidence of different Salmonella serovars in chicken and chicken products from both local and imported source, therefore, 152 samples collected from frozen chickens and chickens products were examined by SMT. Salmonella isolation revealed a total percentage of 5.92% from imported and local frozen chicken and chicken products. The results observed in Table 4 revealed a high incidence of Salmonella serovars isolation among chicken cuts (10%), followed by local frozen chickens and imported frozen chickens samples with incidence of 7.89 and 4.83%, respectively. On the other hand, the whole chicken eggs revealed negative results for isolation of Salmonella species by SMT. The results of SMT revealed that S. enteritidis was dominating among the recovered Salmonella serovars with incidence of 55.56% (5 out of the 9 strains recovered by SMT), followed by S. typhimurium (two strains) with incidence of 22.22%. While the other two strains were S. agona and S. newport. These results indicated the health hazard of poultry as a major source of Salmonella foodborne pathogens (Altekruse et al., 1999;Humphrey, 2002;Schlundt, 2002). The S. agona and S. newport isolated from imported frozen chickens only are indications of the ability of imported chickens to introduce different Salmonella species to the local area that can cause new and devastating outbreaks (Altekruse et al., 1999). The SMT used in these study reported by ISO 6579 (2002) was characterized by very good analytical parameters which allow the detection of low numbers of potentially stressed cells of various Salmonella serovars through the use of pre-enrichment in BPW followed by selective enrichment in RV broth and finally plated into three different Salmonella selective agars.
Traditional methods of identification of food-borne pathogens, which cause disease in humans, are time-  consuming and laborious although control of the infection depends increasingly on the availability of rapid and precise diagnostic tests for monitoring. Therefore, the present study was aimed to investigate the sensitivity of PCR protocol in conjunction with selective enrichment in Rappaport Vassiliadis broth and compared with standard microbiological techniques using field chicken and chicken products collected from Riyadh, KSA. In the present study, the PCR produced positive amplification of 284 and 274 bp fragments of invA and sdiA genes (100%), specific for all members of Salmonella species, respectively, while all non Salmonella serovars (100%) were negative (Figures 1, 2a and 2b). These results were parallel to those obtained by Oliveira et al. (2002) and Malorny et al. (2003), who reported that 139 -141 primers, which target the invA gene were able to identify all the examined Salmonella serovars, whereas all non Salmonella serovars gave negative results. Our results concluded that, all Salmonella carry the invA gene, which is not carried by any other bacterial species (Lin et al., 2007). Regarding detection at the serovar level, the PCR assay for the identification of S. typhimurium was very specific because it could amplify 620 bp fragments of fliC gene in all standard S. typhimurium strains, two isolated S. typhimurium and in none of the other Salmonella serovars (Figure 3a). These data support the work of Oliveira et al. (2002) who correctly identify all S. typhimurium strains but none of the strains from other Salmonella serovars. Moreover, PCR amplify 480 bp fragments of sefA gene in all standard S. enteritidis strains and 5 isolated strains of S. enteritidis; whereas, DNA from other Salmonella serovars and other bacterial genera did not produce any amplification product (Figure 3b). These results confirm the results of somatic serogrouping with polyvalent antisera. Moreover, PCR has several advantages over the slide agglutination test with polyvalent antisera, because serogrouping is not possible when Salmonella isolates lack O-antigen (rough strain) or lack both O and H antigen (Hoorfar et al., 1999). The specificity and sensitivity of PCR assay combined with Rappaport-Vassiliadis (RV) selective enrichment broth (PCR-RV) for the detection of Salmonella species in the collected field samples were tested in this study. The same positive samples detected using SMT were also detected by PCR-RV. In addition, 2 excess positive results only were obtained by PCR-RV assay with a percentage of 5.92 and 7.23% for SMT andPCR-RV, respectively (Tables 5 and 6). The recorded results confirmed that the PCR-RV assay could detect more positive samples of Salmonella species than SMT; these results also confirm the finding of Oliveira et al. (2003) that the PCR test combined with RV selective enrichment is more sensitive in detecting Salmonella at genus level than bacteriological methods.
In conclusion, the PCR assay unequivocally proved to be a highly specific, sensitive and time saving method for detecting Salmonella. The combination of a routine PCR test in conjunction with SMT could be effective in providing a more accurate profile of the prevalence of Salmonella in poultry and poultry related samples.