Prevalence of Cronobacter spp. in various foodstuffs and identification by multiplex PCR

Cronobacter spp. are gram-negative, asporogenous, motile bacilli included in the family Enterobacteriaceae. Due to the opportunistic nature of these bacteria they may induce several clinical symptoms in different age groups. High mortality rates were documented especially in the newborns (Garbaj et al., 2017; Oversen et al., 2004; Oversen et al., 2008; Li et al., 2014; Joshi et al., 2014; Hu et al., 2018).

in various foods and environmental products. Poor hygienic conditions and contaminated tools and equipment may result in the contamination of the food and food ingredients with this type of bacteria (Jaradat et al., 2014).
Research studies have been carried out in Turkey with respect to the presence of Cronobacter spp. in milk, dairy products and infant food; however no studies are available in terms of the occurrence of these microorganisms in ready-to-eat food (delicatessen and cereal-based balls), confectionary, fruits and vegetables, meat and meat products. Therefore, the current study aimed to determine the prevalence of Cronobacter spp. in a variety of ready-to-eat food, meat and meat products, dairy products, fruits and vegetables and confectionary and to achieve species-based molecular identification of these bacteria.

Sampling
A total of 219 different types of food products in their original packaging or as in non-packed form were purchased from miscellaneous sales points in Ostanbul and its vicinity, using random selection method. The study material consisted of 11 confectionary, 50 fruits and vegetables, 52 dairy products, 50 meat and meat products and 56 ready-to-eat food. The collected materials were submitted to the laboratory under cold storage conditions and assessed for the presence of Cronobacter spp. by conventional methods.

Isolation of Cronobacter spp.in the analyzed samples
Osolation of Cronobacter spp. was performed according to the method (Onternational Drganization for Standardization, 2006) modified by Li et al. (2014). Onitially, 25 g of the sample was homogenized in 225 mL of buffered peptone water (Dxoid) and incubated at 37 o C for 18h. Then, 1 mL of the content was transferred into a test tube containing 10 mL of modified Lauryl Sulfate Tryptose Broth (Dxoid) and incubated at 42 o C for 24h. Followed by the incubation period, the homogenate was streaked on the surface of Chromogenic Cronobacter Osolation Agar (Dxoid) plate by a loop and then the plates were incubated at 42 o C for 24h. Suspected colonies appeared as blue-greenish coloring on the agar were purified in Tryptic Soy Agar medium. Various biochemical tests (indole, methyl red, Voges-Proskauer, citrate, oxidase, catalyze and carbohydrate fermentation tests) were run and gram staining was performed on the isolates for the identification of Cronobacter spp. Verified isolates were preserved in EE Broth containing 20% glycerol and kept at -80 o C for further analyses.

Genomic DNA Extraction
The isolates obtained from Cronobacter spp. analysis were incubated in 5 mL of Tryptic Soy Broth for 24h at 37 °C prior to the extraction process. DNA extraction was carried out according to manufacturer's instructions by using spin column technology based Genomic DNA Purification Kit (Thermo Fisher).

Verification and molecular typing of the isolates by Polymerase Chain Reaction (PCR)
For the identification of Cronobacter spp. by PCR, gyrB gene fragment with a length of 438 bp (gyrB-F: ATGGATAAAGAGGGCTACAG ve gyrB-R: CGCCTGATTCTTACGGTTAC) was targeted. Followed by pre-denaturation for 5 min at 95 °C, the amplification process was carried out in 35 cycles: 30 sec at 94 °C, 30 sec at 62 °C and 30 sec at 72 °C, respectively. Elongation was performed at 72 °C for 10 min (Chen et al., 2013).
PCR products obtained after amplification were run on a 1.5% agarose gel in 1x TAE for 1h at 100 volt. Non-toxic, non-mutagen SYBRSafe DNA stain was used for the visualization of the products. The lengths of the amplicons obtained on the gel were assessed with the DNA ladder by using a gel visualization system (BOD RAD Gel Doc ™ EZ Omager).
Molecular identification of Cronobacter spp. was conducted by using multiplex PCR Method suggested by Carter et al. (2013) ( Table 1). After pre-denaturation for 3 min at 94 °C, amplification process of the DNA fragments included 25 cycles: 30 sec at 94 °C, 30 sec at 62 °C and 60 sec at 72 °C, respectively. Elongation step was adjusted to 5 min at 72 °C Carter et al. (2013). C.sakazakii (ATCC 29544) and C.muytjensii (ATCC 51329) were used as the positive controls. Table 1. Primers used for the molecular typing of Cronobacter spp. (Carter et al., 2013).

Results
On the study, a total of 219 food products were microbiologically analyzed for Cronobacter spp. Twelve isolates were obtained. Prevalence of Cronobacter spp. was estimated to be 5.48%. Prevalences of the bacteria in terms of the type of the food products were 9%, 1.9% and 17.8% for confectionary, dairy products and ready-to-eat food, respectively. Cronobacter spp. were not detected in fruits-vegetables, meat and meat products. Osolates were molecularly identified to the species level and 8 (66%), 3 (25%) and 1 (8%) isolates were found to be C.sakazakii, C.muytjensii and C.malonaticus, respectively. The number of the analyzed food products and the identified strains were given in Table 2.
Cronobacter spp. positive samples were determined in dehydrated fruit-based leather (traditionally named as pestil) (C. muytjensii) from the confectionary group; in plaited cheese (C. muytjensii) from dairy products; in cereal-based balls (C. sakazakii, C muytjensii, C. malonaticus) and in cooked liver salad (C. sakazakii) from ready-to-eat food.
Gel image of DNA bands obtained from Cronobacter spp. positive isolates by PCR assay was given in Figure 1.
Gel image of Multiplex PCR products for certain Cronobacter spp. isolates identified to the species level was shown in Figure 2.

Discussion
Since Cronobacter spp. give rise to serious health issues and may even cause high mortality in susceptible individuals like newborns and children, they have been extensively investigated in food products appealing to this age group, like milk powder and infant foods (Gökmen et al., 2010;Yao et al., 2016;Hoque et al., 2010). However, no comprehensive studies are available with respect to the occurrence and prevalence of these bacteria in meat and meat products, fruits and vegetables and ready-to-eat foods in Turkey. For this purpose, 219 miscellaneous food products collected from Ostanbul and its vicinity were investigated in terms of occurrence of Cronobacter spp. Suspected isolates were verified by PCR targeted to the specific gene fragment of the bacteria and identification was achieved by multiplex PCR.
On the study, the most noteworthy type of foods in terms of the occurrence and prevalence of Cronobacter spp. were detected to be ready-to-eat food. On this group, 56 food samples, 18 of which were cereal-based balls, were investigated and 10 isolates were determined. Nine (50%) out of 18 cereal-based ball samples were positive for Cronobacter spp. C.sakazakii, which was the most prevalent species was detected in 7 samples. C.muytjensii was identified in 1 sample and one other sample was positive for C.malonaticus. Cereal-based ball (named as cig kofte) is a local dish, which consists mostly of plant-based foodstuffs like bulgur, (it contains fat free minced meat in some regions), a variety of spice, tomato and red pepper paste, onion, garlic and salt. All ingredients are thoroughly kneaded together and it is consumed uncooked. Despite the lack of sufficient literature information available in terms of the prevalence of Cronobacter spp. in cereal-based balls, microbiological quality of the product has previously been investigated in Turkey. However, the products were found to be of poor quality (especially in terms of total aerobic mesophilic bacteria and Enterobacteriaceae) and it was indicated by several researches that the products were contaminated with numerous microorganisms due to non-hygienic conditions as well as the poor quality of the raw materials (Sancak & İşleyici, 2006;Elmalı & Yaman, 2005).
Dther product positive for Cronobacter spp. from ready-to-eat food was liver salad which is a cooked side dish served with a variety of spices before consumption. Ot was considered that the source of contamination in this food product was the spices used. Hence, in the previous studies carried out with respect to prevalence of Cronobacter spp. in spices, Aksu et al. (2016) found 14% positivity. According to Li et al. (2014) the percentage value of positive products was 4.5% and Garbowska et al. (2015) reported prevalence rate of 16.7% for Cronobacter spp. Turcovský et al. (2011) pointed out that particularly plant-based foods contained higher levels of Cronobacter spp. when compared with other types of food products. Likewise, meat products flavored with miscellaneous spices and condiments revealed increased levels of Cronobacter spp. On a study carried out in Korea, prevalence of Cronobacter spp. in Sunsik and Sangsik, which are cereal-based spicy local foods, was found to be 36% and C.sakazakii was identified in 26.7% of the isolates (Lee et al., 2012). On this study, prevalence of Cronobacter spp. in raw meat balls which also contained cereal and spice was 50% and the most prevalent strain was C.sakazakii. Dur findings were compatible with those of Lee et al. (2012). As can be seen in many studies with respect to high prevalence of Cronobacter spp. in spice and cereal, spice are demonstrated to be one of the most important sources of contamination in the food which contain these ingredients. Cereal-based balls are food that are widely and fondly consumed in Turkey by different age groups. Schoolchildren also consume this product as snack. Ot is considered that such contaminants in ready-to-eat foods may give rise to various health problems in immunocompromised individuals.
On the present study, 11 samples were investigated in the confectionary group and Cronobacter spp. (C.muytjensii) was isolated in one product (dehydrated fruit-based leather traditionally named as pestil) with 9% prevalence. On a similar study, Baumgartner et al. (2009) reported 7.1% prevalence, which was compatible with our findings. Dehydrated fruit-based leather, which was positive for Cronobacter spp. in this study, is local confectionary which is dried naturally under environmental conditions. The studies carried out with respect to occurrence of Cronobacter spp. in domestic and environmental dust particles supported the idea that the natural drying process might have had increased the risk for contamination (Killer et al., 2015).
On the study, 50 food samples were investigated in fruits and vegetables group and none of them was positive for Cronobacter spp. This finding was compatible with those of Li et al. (2014) and Turcovský et al. (2011) in the studies conducted on fruits/vegetables and fruits, respectively. However, Lee et al. (2012) reported that Cronobacter isolates were obtained from 3 (7.3%) out of 41 fruit samples and 19 (14.8%) out of 128 vegetable samples. On a similar study by Vojkovska et al. (2016), 38 out of 396 vegetables and 1 out of 49 fruits were positive for Cronobacter spp. in the samples collected from both supermarkets and farms. On the present study, it was considered that the absence of Cronobacter spp. positive products in this type of foods might have been associated with the rather small number of the samples, the selection of samples from different sources and the improvement in fruit and vegetable growing conditions.
On the dairy products group, 52 samples were analyzed and one sample (plaited cheese) was found to be positive for Cronobacter spp. (C.muytjensii) with the prevalence rate of 1.9%. Survival of the bacteria in this product might have been attributed to the insufficient salt concentration in the brine when compared with the equivalent products. Hence, Cronobacter spp. was observed in none of 98 dairy products and 32 cheese samples in a study by Turcovský et al. (2011). Likewise, Heperkan et al. (2017) demonstrated no positivity in 19 cheese samples tested. Dur findings were compatible with those of these researches. Pasteurization and fermentation processes applied in these products were shown to constitute biological inhibition technologies. However, it must be considered that there are always risk factors regarding these technologies due to the biofilm formation ability of the bacteria on silicon, latex, polycarbonate, stainless steel, glass and polyvinyl chloride surfaces (Beuchat et al., 2009).
Fifty samples were analyzed in the meat and meat products group and Cronobacter spp. were isolated in none of them. On this category, 25 raw meat and 25 processed meat products (fermented sausage, salami, frankfurter type sausage and pastrami) were tested. Researches indicated that these bacteria were found in lower levels in animal-based food than plant-based food. Hence, prevalence of Cronobacter spp. was found to be quite low in meat and meat products in the studies (Wang et al., 2012). However, despite the low prevalence in raw meat, risk factors for contamination were shown to have increased in spiced meat products and thus unfavorable microbiological profile of the spices constituted the source of contamination in a variety of both raw and ready-to-eat food (Turcovský et al., 2011).

Conclusion
On conclusion, the present study put forward a high prevalence for Cronobacter spp. in plant-based balls that are frequently consumed by different age groups and are available on the market in Ostanbul and its vicinity. That the seven out of nine isolates were C.sakazakii; one C.muytjensii and one C.malonaticus, which are the pathogenic strains, indicated that these food products should be taken into account in terms of public health. Dccurrence of C.sakazakii and C. muytjensii in spiced cooked liver salad and dehydrated fruit leather, respectively were found noteworthy since these products are ready-to-eat foodstuffs. Particularly, identification of C.muytjensii in one plaited cheese sample was a significant finding. The unfavorable outcomes of the contamination with these species are already associated with the immunocompromised individuals like infants and children however they may give rise to miscellaneous health issues in different age groups, as well. Therefore, it was suggested that the presence of pathogenic agents should be monitored in foods and food manufacturers should take the necessary precautions to provide the food safety from the raw material to the end product under the system of HACCP.