Many diseases are transmissible via milk products; raw or unpasteurized milk has been a major vehicle for transmission of pathogens (Vasavada, 1988). The use of unclean teats, milking and transporting equipment contributes to the poor hygienic quality of traditional milk products (Altug and Bayrak, 2003). Milk products such as yoghurt, ice cream, and cheese are widely consumed and market for them has existed in many parts of the world for many generations. There are 16 million annual cases of typhoid fever, 1.3 billion cases of gastroenteritis and 3 million deaths worldwide due to Salmonella (Bhunia, 2008). Salmonella is the most frequent pathogen associated with outbreaks of foodborne illnesses (80.5% of the outbreaks), followed by Escherichia coli (10.1%) (De Buyser et al., 2001). Cooked food products and raw milk were most commonly contaminated with food borne pathogens and many of them were resistant to different antibiotics (Walsh et al., 2005; Normanno et al., 2007; Okpalugo et al., 2008 and Novakova et al., 2010). Contamination of dairy foods with virulent pathogens renders them to be a source of public health hazard. The possible contamination sources are either mastitis in dairy cow or the milk itself (Carter, 1995). Ubiquitous serotypes such as Salmonella Enteritidis or Salmonella Typhimurium which affect both man and animals generally cause gastrointestinal infections usually less severe than enteric fever. However, they also have the capacity to produce typhoid-like infections in mice and in humans or asymptomatic intestinal colonization in chickens (Velge et al., 2005). Growing concerns over food safety among the consumers call for the manufacturing and processing of foods under extremely hygienic conditions to avoid possible health challenges (Farzana et al., 2009). According to Pui et al. (2011), epidemiologic classification of Salmonella is based on the host preferences. The first group includes host-restricted serotypes that infect only humans such as S. Typhi. The second group includes host-adapted serotypes which are associated with one host species but can cause disease in other hosts serotypes such as S. Pullorum in avian. The third group includes the remaining serotypes. Typically, S. enteritidis, S. typhimurium and Salmonella Heidelberg are the three most frequent serotypes recovered from humans each year (Gray and Fedorka-Cray, 2002; Boyen et al., 2008). Kauffmann-White scheme classifies Salmonella according to three major antigenic determinants composed of flagellar H antigens, somatic O antigens and virulence (Vi) capsular K antigens. More than 99% of Salmonella strains causing human infections belong to Salmonella enterica subspecies enterica. The infectious dose of Salmonella depends upon the serovar, bacteria strain, growth condition and host susceptibility. However, single-food-source  outbreaks  indicate  that  as  little as 1 to  10 cells can cause salmonellosis with more susceptibility to infection (Yousef and Carlstrom, 2003; Bhunia, 2008). Pui et al. ( 2011) in their review article implicated poultry, eggs and dairy products as the most common vehicles of salmonellosis; including contaminated environment with moving animals (swines, cows and chickens) as vectors hence, constituting important risk factor for infection. Salmonellosis continues to be a major public health problem worldwide and contributes to negative economic impacts thus, making research on Salmonella gained great interest and concern from scientists.

The present study reported the molecular characteri-zation of Salmonella serotypes in cow raw milk, cheese and traditional yoghurt samples in Nigeria and their consequent health hazards in man.

Collection of samples

Thirty samples comprising each of cow raw milk, yoghurt and cheese were collected. Raw milk samples were collected from individual cows marketed at different markets in Ile-Ife, Modakeke, Edun-abon, and Akinlalu, Osun State, Nigeria. The raw milk was collected in a sterile container by placing the container underneath the cow for milking. Yoghurt samples were bought from the hawkers at various markets within the same localities while local cheese (wara) was equally purchased from the same market as above in transparent polythene bags. All cheese and yoghurt samples were transported to the laboratory under refrigeration (4 to 6°C) in thermal boxes containing ice packs and were analyzed bacterio-logically within one hour of collection. Raw milk samples were transported to the laboratory at ca.4°C. All samples were collected between June and August, 2011.

Isolation and characterization of Salmonella isolates

Isolation of Salmonella species was done on MacConkey agar and Salmonella-Shigella agar (SSA) (Oxoid Ltd., Hampshire, England) plates. One milliliter of a five- fold dilution of raw milk and yoghurt was cultured on the sterile agar plates by pour plate technique. For cheese sample, 10 g was aseptically transferred into a sterile mortar and pestle, and homogenized in 90 ml sterile distilled water using stomacher. One milliliter of the 10 fold dilutions of the homogenized cheese was cultured appropriately as above. All plates were incubated at 37°C for 48 h. The isolates were differentiated first on the basis of colonial morphology and microscopic examination. The identity of isolates was confirmed by various biochemical tests with reference to Bergey’s Manual of Determinative Bacteriology (Buchanan and Gibbons, 1994).

Antibiotic susceptibility

Antibiotic susceptibility of isolates was performed by antibiotic disc diffusion technique on Mueller Hinton agar (Hi Media, Vadhani, India) (Clinical and Laboratory Standards Institute (CLSI) 2008). The antibiotics (oxoid, UK), augmentin (30 µg), nitrofurantoin (200 µg), ofloxacin (5 µg), tetracycline (30 µg), gentamicin (10 µg), amoxicillin (20 µg), cotrimoxazole (25 µg), and nalidixic acid (30 µg) were  firmly  placed on the agar plates  previously seeded  with  the test organisms and incubated at 37°C for 24 h. Sensitivity of the isolates to different antibiotics was indicated by the clear zones of inhibition which were measured in milliliter using a calibrated ruler. The diameters of zone of inhibition were compared with the interpretative chart of zone sizes of susceptibility to antibiotics (CLSI, 2008). Resistance to three or more antibiotics by isolate was defined as multiple antibiotic resistance (MAR).

Polymerase chain reaction and amplification of genes

The target genes selected for Salmonella sp. were invasive- invA (Salmonella-specific), adherence- rfbJ(B) (serogroup B), fliC (l, v) and fljB (e, n and z15) using appropriate primers. Oligonucleotide primers were used for the amplification of invA, rfbJ(B) and fljB genes.

A 50 μl reaction mixture contained 1 μl of template DNA (approximately 50 ng), 25 Pmol of each primer, 200 μM deoxyribonucleotide triphosphate mixture, 8 μl of 10X PCR buffer, 1.5 mM MgCl₂, 2.5 U of Taq polymerase was used. The standard cycling condition after the initial denaturation step of 5 min at 94°C, were 1 min at 94°C, 1 min at 57°C, and 2 min at 72°C for 35 cycles and final extension step of 5 min at 72°C. For some strains, the annealing temperature was reduced to 54 to 55°C to increase the amount of product that was amplified. PCR products (10 μl ) were separated by electrophoresis in 0.8% agarose gel in Trisborate buffer (0.089 M Tris-base,0.089 M boric acid, 2.5 mM EDTA-Na₂, pH 8), with the 1 kbp DNA ladder as a molecular size marker.

Statistical analysis

Comparisons of the final values were compared statistically using Chi-square Tests of the statistical analysis system software (SPSS 16.0 version). P value of < 0.05 was considered statistically significant for all the comparisons.

Salmonella sp. were recovered in all the cheese and yoghurt samples but found in only one of the raw milk samples analysed. The percentage occurrence of Salmonella sp. in the samples includes raw milk (10 %), yoghurt (90%) and cheese (100%) (Table 1). The antibiotic susceptibility profile of the isolates from cow raw milk, yoghurt and cheese samples is shown in Table 2. 

Generally, all the Salmonella serotypes were sensitive to ofloxacin and nalidixic acid. Resistance to antibiotics was in varying proportions with 95.2% developing resistance to amoxicillin, 90.4% to tetracycline and 28.5% to augumentin. Resistance was generally higher to amoxicillin than other antibiotics (p>0.05).

Ten (10) different multiple antibiotic resistance (MAR) patterns were displayed by the isolates with 41.17% showing resistance to three different classes of antibiotics (Table 3). All the Salmonella serotypes were multiple antibiotic resistant.

Table 4 shows the result of the polymerase chain reaction for the detection of Salmonella serotypes. The serotypes identified include S. typhi, S. typhimurium, S. Enteritidis, and the serogroups were D, B, D₁ respectively. Out of the 21 strains of Salmonella species, 52.4% belong to B serogroup while 20.0 and 28.6% belong to D and D₁ serogroups respectively. The invA gene was amplified in all the Salmonella serotypes while rfbJ was detected only in S. Typhimurium. The genes fliC and fliB were absent in all the serotypes. Figure 1 presents the PCR amplification of invA and rfbJ genes in Salmonella serotype obtained from cow raw milk and milk products.

All the Salmonella serotypes (S. typhi, S. typimurium and S. enteritidis) were resistant to more than three classes of antibiotics used. Resistance by isolates to antibiotics may however, have been acquired from over exposure to antibiotics particularly in the veterinary sector or to a lesser extent from natural sources especially when some of the antibiotics tested in this study are commonly used in both the health and veterinary sectors. Studies had reported that of the more than 1 million tons of antibiotics released into the biosphere during the past years, approximately 50% are estimated to flow into the veterinary and agricultural channels (Mazel and Davies, 1999). The application of these antibiotics at sub-thera-peutical levels for increased growth and feed efficiencies in farm animals is an integrated part of modern agriculture worldwide, and this leads to the emergence of antibiotic-resistant microbes (Prescott, 2000). When the same antibiotics are used in humans as in animals, resistant microbes which passed on to humans from animal sources could then lead to treatment failure in humans with serious public health implications. The present study recorded high resistance to amoxicillin, cotrimoxazole and tetracycline, which could be due to misuse of these antibiotics in disease conditions. This study also reveals that all the multi-resistant strains showed a characteristic tetracycline resistance trait. This may account for the indiscriminate use of the antibiotic invariably resulting to selection by the organism. Reduced susceptibility to tetracycline has been reported common in S. typhimurium isolates from healthy breeder and broiler flocks in Portugal (Clemente et al., 2014). The multiple resistance recorded in our study was against those antibiotics frequently employed in public health and veterinary sectors. The sensitivity of all the Salmonella serotypes to nalidixic and ofloxacin may account for the non- abuse of these antibiotics.

Pui et al. (2011) in their review reported that Salmonella strains resistance to one or more antibiotics have increased in the Saudi Arabia, United States, United Kingdom and other countries of the world. This is due to the increased and uncontrolled use as well as easy accessibility to antibiotics in many countries of the world (Grob et al., 1998; Yoke-Kqueen et al., 2007). In this study, S. typhi and S. typhimurium were multiresistant to the antibiotics tested, thus confirming the earlier reports in Africa. Emerging resistance in S. typhi has been described especially in Africa and Asia and the appearance of S. typhimurium DT104 in the late 1980s raised main public health concern, thereby threatening the lives of infected individuals (Grob et al., 1998). Evidence exists to suggest that not only are such antibiotic resistant strains more difficult to control in terms of human infection, they may also be more resistant to heat processes (Davidson and Henson, 1995). This is of great concern because majority of infections with MAR Salmonella are acquired through the consumption of contaminated foods of animal origin such as swines and chicken eggs.

Asai et al. (2010) mentioned that cephalosporin and fluoroquinone-resistant strains of S. Choleraesuis have been identified in swines in Taiwan and Thailand. Previous study by Prapas et al., (2008) reported that S. typhimurium and Salmonella Heidelberg ranked first and second, respectively in multidrug resistance, and is among the most commonly isolated serovars from dairy products.

Due to the use of antibiotics for the promotion of growth and prevention of disease in food animals, there is an increase of human salmonellosis cases caused by food-borne MAR Salmonella nowadays (Yang et al., 2010). This indiscriminate and injudicious use of antibiotics in any setting especially in food animals worldwide should be monitored to reduce the transfer risk of MAR Salmonella to humans. Finally, there is a need for continuous surveillance and sharing of antimicrobial susceptibility data for Salmonella among countries worldwide to ensure the effectiveness of control programmes.

The amplification of invA and rfbJ genes by polymerase chain reactions in the Salmonella serotypes revealed that the isolates are pathogenic. S. typhimurium was the most predominant occurring serotype (52.4 %). The presence of invA and rfbJ genes may cause Salmonella infections to become invasive and result in bacteremia and serious extra intestinal disease since it is the genes code for invasion and attachment. Salmonella serotypes have been implicated in several diseases, including enteric or typhoid fever (primarily S. typhi and S. paratyphi), bacteremia, endovascular infections, focal infections (osteomyelitis), and enterocolitis (typically S. typhimurium, S. enteritidis, and S. heidelberg) (Brenner et al., 2000).

The recovery of MAR pathogenic Salmonella serotypes in cheese and yoghurt samples in the study areas calls for great health concern as these organisms have been associated with salmonellosis and other grave diseases. Good quality raw materials for product processing, adoption of Good Manufactured Practices (GMP) and strict personal hygiene will ensure safety and high quality dairy products.

The author(s) have not declared any conflict of interests.

Authors thank Mr. A. George, University of Melbourne, South Africa for the assistance rendered in the molecular aspect of the work.