Molecular typing and cytotoxicity testing of Staphylococcus aureus isolated from milk , meat and clinical sources

Center of Excellence in Post-Harvest Technologies, North Carolina Agricultural and Technical State University, the North Carolina Research Campus, Kannapolis, NC 28081, North Carolina, United States. Department of Food and Animal Sciences, Alabama Agricultural and Mechanical University, Food Microbiology and Immunochemistry Laboratory, Normal, AL. 35762, United States. Department of Food Science and Technology, Bowen University, Iwo, Osun State, Nigeria


INTRODUCTION
Staphylococcus species are found normally in hair and skin of humans and animals, and according to many researchers, they are currently the leading cause of nosocomial infections worldwide (Lowy, 2003;Michelim et al., 2005;Bayles et al., 1998).Staphylococci, espe-cially Staphylococcus aureus, are opportunistic pathogens if infection occurs in the susceptible individuals (Ghebremedhin et al., 2007).
Infections associated with S. aureus are extremely common, ranging from furuncles to foodborne intoxication and life threatening disease; therefore, the pathogen has potential to cause increased morbidity and mortality (Bayles et al., 1998).A recent study shows that almost 50% of meat in the grocery stores in the United States is contaminated with S. aureus (Waters et al., 2011); it is also the leading cause of intra-mammary infection in dairy animals.Furthermore, comprehensive epidemiological analysis has shown that Staphylococcus spp. was involved in several cases of outbreaks in United States (Ajayi et al., 2011).
According to reviews in Food Science and Food Safety (2006), there are antimicrobial usages at several stages in the food system, from the environment into the kitchen table, which creates an increasing spread of multi-drug resistant S. aureus in food worldwide and is a threat to food safety creating a public health concern (Xie et al., 2011).Because of the spread of multi-drug resistant pathogens in foods, the Staphylococcus spp. was subjected to resisto typing, applying 14 classes of antibiotics in order to establish the antimicrobial resistance profile of the isolates.It was found that majority of the isolates were resistant to 13 out of the 14 antibiotics tested (Ajayi et al., 2009).
Additionally the study on virulence genes characterization of Staphylococcus spp.isolated from food and clinical sources is important to estimate the prevalence of infections.Pulse-field gel electrophoresis (PFGE) is known to be a highly discriminatory technique, often used for characterizing genetic diversity of isolates and to research on outbreak sources (Xie et al., 2011).Furthermore, PFGE is particularly effective for molecular typing of methicillin resistant S. aureus (MRSA) (van Belkum et al., 1998).
The persistence of staphylococci in its hosts, despite the initiation of humoral and mucosal antibodies, continues to be an important issue for health of both humans and animals (Bayles et al., 1998).Furthermore, bacteria such as Listeria monocytogenes, Salmonella spp.and Shigella spp.are known to evade humoral immunity and then become internalized in host cells (Bayles et al., 1998).Although S. aureus is generally not considered to be an intracellular pathogen, it can be internalized, inducing apoptosis (cell death) in various cell types including epithelial cells, endothelial cells, keratinocytes, osteoblasts, lymphocytes and macrophages (Bayles et al., 1998;Wesson et al., 2000;Menzies and Kourteva, 1998;Nuzzo et al., 2000;Tucker et al., 2000;Jonas et al., 1994).The human INT-407 cell line is used in pathogenic enterobacterium studies, and to study cytotoxicity potential of foodborne pathogen Escherichia coli, (Maldonado et al., 2005), Salmonella enterica serovar Typhi (Santos et al., 2001) and Klebsiella pneumoniae.
Lactate dehydrogenase (LDH) is a low molecular weight enzyme, easily released from cell as a result of minor disturbances to the cell membrane integrity (Bhunia, 2008), and the release of LDH is one of the most widely used and accepted methods for measuring cellular death/lysis (Moran and Schnellmann, 1996).One of the advantages of this assay is that it can detect minimal membrane damage.Therefore, because of the prevalence in nature and the role of S. aureus in foodborne diseases and nosocomial infections, the focus of this study are to examine the virulence genes using multiplex PCR and the genetic relatedness of the strains using PFGE.Secondly, we aimed to determine the cytotoxic potential of strains isolated from milk, meat and clinical sources.

Isolation of S. aureus from clinical, rabbit meat and milk samples
The clinical samples were donated from a local hospital and the milk samples were collected weekly from a local diary (Huntsville, AL) and processed in the Food Microbiology Laboratory at Alabama A&M University.Isolation of S. aureus was conducted according to the procedure described by the FDA-Bacteriological Analytical Manual (FDA, 1998).For initial isolation of Staphylococcus in meat (Rabbit), one hind leg from each whole carcass was placed in a sterile Stomacher bag with 0.1% (w/v) peptone (Oxoid, Lenexa, KS) water and vigorously shaken for 2 min and further dilutions were prepared.Approximately 0.1 ml of each dilution was plated onto Baird-Parker agar (BPA) supplemented with egg yolk tellurite (Merck, Gibbstown, NJ) and incubated at 37°C for 24 h.For isolation of S. aureus from milk or clinical samples, a 0.1 ml of both undiluted and 1:10 serially diluted sample of each was plated and incubated as earlier described.
Plates containing colonies with typical growth morphology of S. aureus were selected, isolated and sub-cultured in trypticase soy broth (TSB).Staphylococcal isolates were identified by colony morphology, catalase, Gram-stain, Staphylococcal coagulase reaction and clumping factor (agglutination) (Food and Drug Administration, 1992).Identification was confirmed with API kits, either API STAPH or ID 32 STAPH (bioMerieux NA, Durham, NC).
S. aureus ATCC 700698, and E. coli which served as the positive and negative controls respectively were included in the PCR reactions.About 5 μL aliquot of PCR products was mixed with 3 μL loading dye and with molecular weight marker of 100-bp, were ana-lyzed by electrophoresis on a 1% agarose gel at 70V for 1 h.Gels were stained in ethidium bromide, viewed by ultra-violet (UV) transillumination and photographed.

Pulsed PFGE
Only Staphylococcus isolates that tested positive for either cflA or mecA virulence genes were subjected to PFGE for research on clonal relatedness.

Preparation of DNA plugs
Chromosomal DNA was prepared as described by Reed et al. (2007).Briefly, 5 μL of lysostaphin enzyme was added to cell suspension equivalent to 3-4 x 10 9 CFU/mL, plus the addition of 250 μL of 60°C 1.8% SeaPlaque agarose (FMC Bio-product).Suspension was mixed well and 250 μL was immediately transferred into the plug molds.Plugs were allowed to solidify for 10 min at 4°C, washed and DNA plugs were cut and placed in an eppendorf tube containing 125 μL of 1X SmaI restriction buffer.Then, 30 U of SmaI restriction enzyme (Roche Diagnostics, Mannheim, Germany) was added to each tube, mixed gently and incubated for 2 h at 25°C.The digested DNA plugs were resolved on a 1% gel on a Chef Mapper (BioRad, Hercules, CA, USA), PFGE equipment with CHEF MAPPER parameters set as follows: 5 s initial switch time; 40 s final switch time; 20 h duration of run; voltage 200 V or 6 V/cm gradient; angle of 120° and chiller at 14°C.All gels were stained with ethidium bromide, documented and analyzed with a Foto/analyst Luminary FX Electrophoresis Documentation and Analysis System (Fotodyne Inc.Rochester, NY).Similarity indices were determined using the Dice coefficient (Dice, 1945) and the distances between clusters were calculated using the unweighted pair group method with arithmetic averages (UPGMA).Data analysis was performed using SPSS version 12.0.1.(SPSS Inc.Chicago, IL).Interpretations of PFGE patterns were performed according to Tenover et al. (1995).

Preparation of mammalian cell culture
The Human jejunal (INT-407) cell line was obtained from American Type Culture Collection (ATCC CCL-6), cultured and maintained in Dulbecco's modified Eagle's medium (DMEM) (Sigma Chemical Co., St. Louis, MO) with 10% fetal bovine serum (D 10 F).Stock INT-407 cultures were maintained in a 37°C incubator with 7% CO 2 for 3 weeks to ensure a uniformed monolayer of culture.Confluent monolayer used for cytotoxicity assay was trypsinized, resuspended in 24-well plates with fresh medium and incubated for 48 h.

Bacterial cultures
Stock Staphylococci cultures isolated from food and clinical sources were maintained in Trypticase Soy Broth (TSB) with 20% glycerol and stored at -80°C.Prior to use, each strain was sub-cultured in 5 mL TSB and incubated overnight at 37°C.

Toxin preparation
Toxin was prepared as described by Maldonado et al. (2005).Briefly, a 1.5 mL aliquot of overnight bacterial culture was centrifuged at 10,000 rpm for 3 min and cell-free supernatants were stored in sterile tubes.Cell pellets were re-suspended in 75 μL of polymyxin B sulfate solution (2 mg/mL of PBS), incubated for 30 min and centrifuged.Following centrifugation, the supernatants were collected, combined with the original cell-free supernatants and filtered through 0.45 µm disc filters (Corning, NY).The filtrates were used immediately or stored at 4°C for up to two days.

Lactate dehydrogenase (LDH) assay
The cytotoxic effect of Staphylococcus on human cell lines was assessed, by measuring lactate dehydrogenase (LDH) released according to Roberts et al. (2001), with slight modification.Briefly, confluent INT-407 cells transferred to 24 well plates were used upon formation of monolayer.Each well was washed three times in 900 μL serum free (D0F) DMEM, then the low (PBS), high (1% Trition-X) controls and samples were prepared in triplicates in a 24 well plates as follows: (900 μL D 0 F DMEM and 100 μL of controls or bacterial toxin filtrates), and incubated for 16 h at 37°C with 7% CO 2 .After incubation, a 100 μL aliquot of supernatant from each bacterial isolate was dispersed into triplicate wells of a 96 well plates.Lactate dehydrogenase substrate was prepared according to the manufacturer's guidelines (Roche Applied Science, Indianapolis, IN) and 100 μL was dispensed into each well.Plates were incubated at room temperature and the absorbance was measured at a dual wavelength of 490 and 655 nm after 5 min and subsequently every 3 min for 15 min by a plate reader (Synergy HT, Vermont, USA).Percentage of cytotoxicity was calculated as follows: Experimental value -Negative control Cytotoxicity (%) = x 100 Positive control -Negative control

Statistical analysis
Calculated percent cytotoxicity data were analyzed using Statistical Package for the Social Sciences (SPSS 16).Averages of the triplicates plus Standard deviation were reported.Tukey's studentized test was used for the separation of means and P<0.05 was taken as statistically significant association.

Multiplex PCR: 16S rRNA; (clfA) and (mecA) genes
Multiplex PCR analysis was performed on 61 samples of presumptive Staphylococcus isolates; 11 from milk; 27 from meat and 23 from human infections (clinical) and S. aureus ATCC 700698, and E. coli which served as the positive and negative controls respectively were tested.PCR amplifications yielded the fragments of the expected sizes.All Staphylococcus isolates were confirmed positive for 16S rRNA, 28 isolates for clfA and 16 strains were positive for mecA gene (Figure 1).
Clumping factor (clfA) gene was found in 20 clinical and eight milk isolates.None of the strains isolated from rabbit meat were positive for clumping factor gene.The mecA gene which resulted in a 499 bp fragment was found in 9 clinical and 7 milk isolates, while none of the meat isolates showed positive mecA gene (Figure 1).The complete results of the presence of 16S rRNA, clfA and mecA virulent genes for each strain are summarized in Table 1.

Clonal relatedness in Staphylococcus
To investigate the clonal relationship, a total of 33 S. aureus strains (9 isolates from milk and 24 from clinical sources) were subjected to SmaI macro restriction fragment analysis of their chromosomal DNA by PFGE, including 2 ATCC (8325-4 and 700698) strains of S. aureus isolates.The PFGE profiles of the clinical isolates were compared with the PFGE profile of the milk isolates.
As reported by Tenover et al. (1995), isolates are designated genetically indistinguishable if the restriction patterns have the same number of bands and the bands appear to be of similar size.Figure 2A is representative of the PFGE gels for clinical strains, two strains in lanes 3 and 4 had PFGE banding patterns that were indistinguishable.
The other clinical strains were closely related, since there were two to three differences in bands between the observed patterns, possibly due to point mutations (insertions or deletions).The S. aureus strains from milk are shown in Figure 2B.Lanes 1, 7 and 15 correspond to ATCC 700698 and lanes 6, 8 and 13 to one milk sample that appeared to be genetically indistinguishable.
Of the 33 isolates typed by SmaI-based PFGE, each of the two ATCC strains had distinct PFGE patterns different from the strains isolated from milk or clinical sources.The horizontal bar above the dendrogram indicates the percentage of genetic relatedness among the different strains.Each cluster had varying number of strains.

Cytotoxicity profile
A total of 57 Staphylococcus isolates were evaluated for their cytotoxicity potential, based on the release of lactate dehydrogenase from jejunal (INT-407) cells, and were compared with the positive control, Triton-X 100 (1%) and PBS and the negative control.Cytotoxicity assay were performed in triplicate.The LDH release values for Triton-X 100 ranged from 1.459 to 1.47 and the negative control ranged from 0.334 to 0.338.The results of LDH values (averages of absorbance ± standard deviations) and calculated percent cytotoxicity for each isolate tested are listed in Table 1.
The strains isolated from milk had LDH had range from -24 to 80%; the strains from meat ranged from -27 to 194%; while the clinical strains LDH ranged from -17 to 113%.There was a statistically significant difference (p<0.05) in LDH release between the clinical and milk  isolates, and also between meat and milk isolates as shown in Figure 5.However, there was no significant difference (p<0.05) in LDH release between the clinical and meat isolates.The meat isolates displayed the widest range of LDH release, followed by the clinical isolates.However, the clinical isolates had the smallest number of isolates with negative LDH release.This assay is used to assess the pathogenic capabilities of the strains isolated from milk, meat and clinical sources.

DISCUSSION
In this study, 61 S. aureus strains isolated from milk, rabbit meat and human infections were investigated for 16S rRNA gene and cflA and mecA virulence genes by PCR.PCR techniques provide increased sensitivity, allow for more rapid laboratory processing times and enhance the likelihood of detecting bacterial pathogens (Riyaz- Ul-Hassan et al., 2008).It has also been proposed for detection of foodborne pathogens to replace the time consuming classical techniques (Olsen, 2000).The strains were also genotyped using PFGE and the cytotoxicity profile was determined.All 61 isolates from this study were positive for 16S rRNA gene, as well as in previous studies (Mason et al., 2001;Zhang et al., 2004;Fan et al., 2008).The 16S rRNA gene is specific for Staphylococcus spp.Clumping factor (clfA) gene, was amplified for 20 of the clinical, 8 milk isolates and none of the strains isolated from rabbit meat.The isolation of clumping factor A is specific to S. aureus.Furthermore, clfA gene codes for a multifunctional protein, in addition to binding fibrinogen and it can also contribute to S. aureus capabilities of binding to platelet (Siboo et al., 2001).S. aureus have also been documented to produce other fibrinogen binding proteins such as clfB, although the primary fibrinogen-binding protein is clfA (Ni Eidhin et al., 1998).Pathogenesis of S. aureus results from the combined action of a diversity of factors, and infection begins with bacterial adhesion to host tissues (Jarraud et al., 2002).The clfA gene amplification results (87%) for the clinical strains in this study are comparable to previous studies by Tristan et al. (2003), who reported that clfA gene was present in 100% of the clinical samples tested while Peacock et al. (2002) reported 98% of clfA gene.Raw unpasteurized milk has been shown to be natural reservoir for S. aureus and other pathogenic microorganisms.In addition, studies have further demonstrated that S. aureus can be isolated from raw milk, 43-62% of the times (Foschino et al., 2002;Chye et al., 2004).From the raw and unpasteurized milk procured for this study, 73% of the S. aureus isolates were positive for clfA gene.The emergence and increasing number of methicillin resistant S. aureus in foods and the environment are alarming.Furthermore, methicillin resistance gene in S. aureus is conserved and its amplification can be beneficial during outbreak investigation.Only 45% of clinical and 64% of milk strains was positive for mecA gene.Previously, it was reported by Suzuki et al. (1993) that in some strains of Staphylococcus the loss of their regulator gene by deletion occurs and we speculate that in this research some clinical strains may have lost the regulator gene also.However, more of the strains isolated from milk had mecA gene.This poses a considerable public health risk, in particular because of the increase in number of individuals who consume raw milk (Riyaz-Ul-Hassan et al., 2008).It is known that the food handlers are most often exposed to S. aureus contamination (Jablonski and Bohach, 2011;Jørgensen et al., 2005) and that raw milk and raw milk products from dairy animals, un-clean tanker and food handlers are probably equally involved as source of the contamination.
All the S. aureus strains from rabbit from this study were negative for the clumping factor A and methicillin resistance genes.According to Vancraeynest et al. (2006), a high virulence (HV) rabbit S. aureus clone causes transmission of chronic staphylococcosis, whereas low virulence (LV) strains only infect a limited number of animals.It is therefore speculated that the Staphylococcus strains from the rabbit isolates could be low virulent strains and that probably explains the inability of PCR to detect either clumping factor A or methicillin resistant gene.
All the strains were positive for 16S rRNA gene in Staphylococcus isolates.The PFGE was used to further discriminate the isolates.As reported by Revazishvili et al., (2006), PFGE is more useful for short-term studying of S. aureus; and is considered the gold standard technique for MRSA typing because of its high discriminatory power, its reproducibility, and its good correlation with epidemiologically linked data (Tenover et al., 1994).The results of the PFGE typing showed some degree of genetic heterogeneity among the S. aureus strains from clinical and milk sources.At 80% similarity level, 8 clusters were identified (Figure 3).The isolates from similar clinical or milk sources were grouped into the same PFGE cluster in this study in contrast to Xie et al. (2011) who reported same PFGE cluster organization from strains of Staphylococcus isolated from different sources.Five MRSA strains from clinical sources were not grouped into PFGE cluster while six strains isolated from milk were in one cluster.
Lactate dehydrogenase is a low molecular weight enzyme carried in the cytoplasm of the majority of cells, which is then released from the cells after a minor disturbance of the integrity of the cell membrane.The LDH results were comparable with LDH release in E. coli, using Vero cell assay (Maldonado et al., 2005), although in this study jejunal INT-407 cell culture was used to access the cytotoxicity of the Staphylococcus strains.The LDH release cytotoxicity assay from human intestinal cell line was applicable to determining the virulence capabilities of Staphylococcus strains isolated from food and clinical sources.Lactate dehydrogenase release also signified the pathogenic effect of Staphylococcus by causing cell damage and cell death of the INT-407.
However, the results of this study contradict Buxser and Bonventre (1981) study, which found that Staphylococcus enterotoxins A and B did not initiate LDH release in INT-407 cells.
Lactate dehydrogenase assay was performed to show the virulence potential of Staphylococcus isolates from milk, meat and clinical sources.According to Bhunia (2008) although S. aureus is generally not considered to be an intracellular pathogen associated with classical facultative intracellular pathogens such as Listeria, Salmonella, and Shigella spp., it has been documented that some S. aureus strains isolated from bovine mastitis can adhere and invade bovine mammary epithelial cell line MAC-T and primary bovine mammary secretory epithelial cells (Almeida et al., 1996;Bayles et al., 1998).

Conclusion
These findings show the prevalence of S. aureus in food and clinical specimen, and we provide evidence that even if virulent genes such as clfA or mecA genes were not amplified in some strains, there may still be cytotoxicity effects presenting a public health concern particularly in individuals that consume game animals or raw milk and might have compromised immunity.

Figure 2 .
Figure 2. PFGE patterns of Sma1 restriction enzyme digested DNA of S. aureus.Lanes 1 to 5 are representative of PFGE pattern from clinical (A) and milk (B) strains; lanes 1, 7 and 15 correspond to ATCC 700698; lanes 2 to 6 and 8 to 13 correspond to milk strains.

Figure 3 .
Figure 3. PFGE macro restriction profiles (SmaI) of S. aureus strains from clinical and milk sources.Pair comparison and cluster analysis performed with the Dice correlation coefficient and UPGMA.

Figure 5 .
Figure 5. LDH release of Staphylococcus strains isolated from milk, meat and clinical sources.Mean separation is by Tukey's studentized analysis, means with same letters are not statistically different (p<0.05).

Table 1 .
Cytotoxicity Profile of Staphylococcus aureus isolated from milk, meat and clinical sources.