Monitoring of selected parasites and other foodborne microorganisms in vegetable products at point of sale

It is supposed that epidemiological data linked to cases of alimentary infection from fresh produce are scarce and usually underestimated. The present study reveals data concerning the occurrence of less studied foodborne pathogens such as parasites ( Giardia intestinalis , Cryptosporidium parvum/hominis ) and viruses (norovirus, hepatitis A virus) in 175 samples of fresh-cut vegetables, frozen vegetables and sprouts. In addition, samples were also analyzed for the presence of bacterial indicators of production hygiene such as Escherichia coli and Staphylococcus aureus . Our results revealed that DNA of G. intestinalis was found in two samples (1.1%) of mixed fresh vegetable salads and mixed frozen vegetables at a level of 100 CFU/g. Norovirus was detected in two samples (1.1%) of rucola and frozen mixed vegetables. No sample was positive for C. parvum/hominis or hepatitis A virus. Our results highlight the need for microbial analysis of food of non-animal origin for specific parasitic and viral agents not generally monitored in these food category types, but their presence poses risks for alimentary infection.


Abstract
It is supposed that epidemiological data linked to cases of alimentary infection from fresh produce are scarce and usually underestimated. The present study reveals data concerning the occurrence of less studied foodborne pathogens such as parasites ( Giardia intestinalis , Cryptosporidium parvum/hominis ) and viruses (norovirus, hepatitis A virus) in 175 samples of fresh-cut vegetables, frozen vegetables and sprouts. In addition, samples were also analyzed for the presence of bacterial indicators of production hygiene such as Escherichia coli and Staphylococcus aureus . Our results revealed that DNA of G. intestinalis was found in two samples (1.1%) of mixed fresh vegetable salads and mixed frozen vegetables at a level of 100 CFU/g. Norovirus was detected in two samples (1.1%) of rucola and frozen mixed vegetables. No sample was positive for C. parvum/hominis or hepatitis A virus. Our results highlight the need for microbial analysis of food of non-animal origin for specific parasitic and viral agents not generally monitored in these food category types, but their presence poses risks for alimentary infection.

Background
According to epidemiological statitstics it has been demostrated that outbreaks of foodborne infections are mostly associated with foods of animal origin. However, a higher number of announced outbreaks and cases associated with food of nonanimal origin has been documented recently [1]. Generally, the most commonly associated pathogens with fresh and minimally processed produce are norovirus (NoV), Salmonella spp. and pathogenic Escherichia coli. Regardless, other foodborne pathogens linked with fresh produce may be recorded. Except for other well known bacterial agents (Shigella sp., Yersinia enterocolitica, Y. pseudotuberculosis, Staphylococcus aureus and Listeria monocytogenes),, viral agents such as hepatitis A virus (HAV) and protozoan agents such as Cryptosporidium spp., Giardia intestinalis (syn. G. lamblia or G. duodenalis),, and Cyclospora sp. have been highlighted in recent years [1,2,3]. Nevertheless, it is supposed that epidemiological data linked to cases of alimentary infections originating from fresh produce are scarce and usually underestimated. For example, underestimation of parasitical infections is caused by the fact that the diseases manifest after long incubation period (average of 7 days) and shelf life of fresh vegetable products is too short to confirm the the parasitic agents in contmainated products. In comparison, infection caused by viral agents such as NoV have short incubation time andlast just a few days (e.g. 24-48 hours for NoV) and people usually heal without seeing a physician. Additionally, physicians do not carry out any identification of the pathogen in majority of cases of intestinal infection.
Furthermore, notification to public health authorities is not compulsory for most parasitic and viral diseases, and therefore official reports do not reflect the true prevalence or incidence of disease occurrences [2].
In our survey, we focused on minimally processed vegetables such as cut vegetables, ready to eat vegetable salads and sprouted seeds that pose a higher risk of foodborne infection as they are consumed without heat treatment. Further, traditional washing with authorized disinfection methods cannot properly eliminate foodborne pathogens from these matrices. A failure in decontamination may be attributed to biofilm formation and incorporation of pathogens on the inner side of plant tissue or higher resistance of some strains to disinfection.
Although information on the prevalence, survival and multiplication of bacterial pathogens on fresh produce is common [4,5,6], there is still insufficient information on the prevalence of foodborne parasites and viruses.
Generally, it is supposed that parasitic diseases are the main issue in developing countries, but due to the cosmopolitan markets with food, changes in climate and agriculture, the increasing number of immunocompromised humans, they may pose a threat to human health in high-income countries as well. In the present work we focused on medically important protozoa Cryptosporidium sp. and G. intestinalis.
Both agents are considered the most important diarrheal pathogens affecting people worldwide. The most commonly affected patients are young children and immunocompromised patients who experience clinical disease more frequently and with higher severity. According to the Striepen [7], cryptosporidiosis is the second most common diarrheal infection in infants under two years in developing countries.
In industrialized countries, reported cases of cryptosporidiosis are not so prevalent, however the number of diagnosed cases are increasing. In the EU, 10 915 cases of cryptosporidiosis were confirmed in 2015, which was 41% more than reported in 2014 [8]. In the same year, giardiasis was confirmed in 18 031 cases with an increase of 4.4% from 2014 [9]. Both infections are transmitted via the fecal-oral route and result from ingestion of cysts/oocysts through the consumption of contaminated food or water or through direct contact with infected humans.
Genotyping and subtyping data suggest that zoonotic transmission is more prevalent in the epidemiology of cryptosporidiosis than of giardiasis.
Norovirus is the most common worldwide cause of non-bacterial gastroenteritis. This virus is highly infectious with an incubation period of one to three days and an infectious dose requires only a few copies of viral particles. In comparison, a long incubation period lasting three to six weeks is observed in HAV infection, which complicates the tracing of contaminated food. Generally, viruses transmitted via the fecal-oral route are highly resistant to cooling, freezing, a wide range of pH and various disinfectants that are effective for bacterial agents. The World Health Organization identified both NoV and HAV in fresh produce including vegetables as a priority virus/commodity combination for which control measures should be considered [10].

Results
The results obtained from the examination of fresh pre-cut produce, frozen vegetables and sprout seeds are summarised in Table 1. Parasites and viruses were each detected in 1.14% of studied samples, separately. G. intestinalis was recoverd from one sample of fresh leafy green mixed salads and from one sample of frozen vegetables. No sample was positive for C. parvum/hominis. Out of studed viruses, only NoV was detected in one sample of rucola and one sample of mixed frozen vegetables. Both NoVs were classified as NoV GI.
S. aureus was isolated from 1.7% of samples (one fresh mixed salad and two frozen peas) at a level less than 50 CFU/g. Based on PCR analysis, genes for production of classical enterotoxins (sea-see) were not confirmed in these isolates. E. coli was isolated from 43.4% of samples, when sprouts seed were the most commonly contaminated category with the highest level of contamination. Fifty-one percent of sprout seeds samples exceeded the level of 10 3 CFU/g and 20.0% of samples exceeded the level of 10 5 CFU/g. In comparison, the contamination level of fresh pre-cut produce by E. coli was only 3.1% with the maximum level of E. coli 3 x 10 3 CFU/g. Surprisingly, E. coli was not present in both NoV positive samples and both G. intestinalis positive samples contained E. coli but only at a level of less than 10 2 CFU/g.

Discussion
Out of 175 examined samples, four samples (2.2%) were positive for parasitic or viral pathogens. Although parasites such as Cryptosporidium spp. and G. intestinalis may be present in produce and can be identified as causes of foodborne diseases [11,2], there are only a few reports concerning their prevalence on produce. Most studies are from developing countries where poor hygiene is prevalent [12,13,14] and from north European countries where outbreaks connected with parasites were described in recent years [11,15,16].
In the present study, the occurrence of parasites in fresh-cut vegetables, frozen vegetables and sprouts was low (1.1%). G. intestinalis was detected in two samples (mixed fresh vegetable salads and mixed frozen vegetables) at a level of 100 CFU/g.
No sample was positive for C. parvum/hominis. Dixon et al. [17] revealed a similar rate of contamination of G. intestinalis (1.8%) in packed precut salads and leafy greens retailed in Canada. On the other hand, the occurrence of Cryptosporidium spp. in Canada was higher, and reached 5.9%. In comparison in Egypt, Gardia spp.
was found in 8.8% and 8.2% of vegetables by Eraky et al. [13] and Hassan et al. [18], respectively. Duedu et al. [12] revealed 6.0% of G. intestinalis and 17.0% of Cryptosporidium spp. in fresh vegetables sold in Ghana. The waterborne route and manure application to vegetable fields is the most important means of transmission of these parasites [16]. In some cases, transmission of these pathogens to vegetables were attributed to infected food-handlers [19].
NoV (NoV GI) was demonstrated in both raw minimally processed (rucola) and frozen vegetables (frozen mixed vegetables) in 1.1% of studied samples. This relatively low rate of NoV contamination of fresh and frozen vegetables is in general agreement with data reported in previous European studies in which NoV was detected in less than 0.1% (1/1372) in fresh leafy samples and in no samples of 1 160 tested ready to eat vegetables [20]. Despite the generally low rate of NoV contamination of food, an increase of foodborne outbreaks caused by NoV has been noted and many outbreaks were traced to food that was handled by infected workers during food preparation. The impact of vegetable contaminated by NoV has been outlined e.g.
S. aureus is part of the natural microflora of humans and therefore the most common way for produce contamination is through direct contact with handlers.
Bacteriological examination of food handlers carried out by Saeed and Hamid [19] revealed S. aureus as the most abundant pathogen. Illness results from an enterotoxin, which is produced by some strains on the food. The true incidence of staphylococcal food poisoning is unknown for a number of reasons, including the short period of illness usually not requiring seeing a physician. In the present study, S. aureus was isolated from 1.7% of samples (one mixed salad and two frozen peas) at a low level less than 50 CFU/g. Genes for production of classical enterotoxins (sea-see) were not confirmed. This result shows a lower prevalence of S. aureus than revealed in Thailand and China [23,24]. Ananchaipattana

Preparation of samples for parasitological and virological analyses
One hundred grams of each sample were placed in a stomacher bag and artificially contaminated with 5 µL of process control virus (PCV, 106 particles/µL) to verify analysis for the presence of viral pathogens [29]. Subsequently, 230 mL of Trisma base-glycine beef extract (TGBE washing buffer, pH 9.5, in house) was added and homogenized for 2 min in order to remove microorganisms from vegetable surfaces.
The liquid was decanted into tubes and centrifuged for 20 min at 8 000 × g at 4°C.
The sediment (pellet) was used for DNA isolation and subsequent detection of parasitic agents by real-time PCR (qPCR) and the supernatant was used for further analyses focusing on the presence of foodborne viruses.
Isolation of parasitic DNA and detection of G. intestinalis and C.

parvum/hominis by qPCR
The pellet was immediately resuspended with 10 mL of BPW; two ml of the resuspended pellet was centrifuged at 14 000 × g for five min. DNA was isolated from the prepared pellet using the PowerSoil DNA isolation kit (MoBio Laboratories, U.S.A) following the manufacturer´s instructions. The protocol was slightly modified to include mechanical homogenization with zirconia/silica beads (0.1 mm) in a MagNA Lyser instrument (Roche, Mannheim, Germany) at 6 400 rpm for 60 s and each sample was supplemented with Carrier DNA solution (salmon sperm DNA, 50 ng/ µL; Serva, Heidelberg, Germany). Isolated DNA was stored at -80°C or directly applied for qPCR. The detection of G. intestinalis and C. parvum/hominis via species specific loci (β -giardin and hsp70) was adopted from Helmi et al. [30] and Centrifugation at 10 000 × g for 30 min at 4°C followed. The supernatant was discarded. A further centrifugation at 10 000 × g for 5 min at 4°C was performed and the pellet resuspended in 1 ml of phosphate buffered saline (PBS), transferred into a clean tube and 1 mL of chloroform: butanol (1:1) was added. The mixture was vortexed, incubated at room temperature for 5 min and centrifuged at 10 000 × g for 20 min at 4°C. The aqueous phase was transferred into a clean tube and retained for nucleic acid isolation, which was done by the NucliSENS Magnetic Quantification of PCV as well as calculation of processing efficiency (estimate of PCV recovery) was done as defined by Mikel et al. [29]. The samples where processing efficiency was lower than 1% were not considered as valid results and not included in further analyses.  Availability of data and materials All data generated and analyzed during this study are included in this published article.