First molecular detection of Toxoplasma gondii in vegetable samples in China using qualitative, quantitative real-time PCR and multilocus genotyping

Toxoplasma gondii infection is becoming increasing problem in China but there is no data concerning contamination of vegetables intended for consumption with this parasite. The aim of the present study was to investigate fresh vegetables originated from open markets located in the Xining City, the Qinghai-Tibet Plateau (QTP), P.R. China for their contamination with T. gondii. A total of 279 fresh vegetable samples were collected and analysed using real-time PCR assay targeting B1 gene and multilocus genotyping. T. gondii DNA was found in 10 (3.6%) samples tested; eight of them represented T. gondii type I and remaining two T. gondii type II. The approximate level of contamination of positive vegetables samples, estimated based on quantitative real-time PCR (qPCR), ranged between less than one and 27000 T. gondii oocysts per sample, with majority not exceeding several oocysts per sample. The results of the study confirmed that T. gondii is present in vegetables offered in open markets in the Qinghai province, P.R. China; eating them unwashed and raw may therefore pose a threat to consumers. This is the first investigation describing T. gondii detection in fresh vegetables intended for consumption collected from the territory of P.R. China using sensitive molecular tools.


Samples and Study Area
Plateau Agricultural Products Distribution Centre, a big hub for agricultural products. Only about 10% of them are grown in the Qinghai Province and the vast majority are transported there from other provinces including: Sichuan, Shandong, Shaanxi, Henan, Yunnan and Gansu provinces (Fig. 1). We were not able, however, to identify the origin of each separate vegetable that was sampled. Vegetables are cultured in greenhouses and farmlands which are located in the suburbs of the cities or in rural areas close to countryside. These places may attract animals including cats being definitive hosts of T. gondii and contamination of plants with oocysts is probable.

Results
In total, 279 fresh vegetable samples originated from open markets in the Xining City, QTP, P.R. China were examined with real-time PCR assays based on the T. gondii B1 gene. The presence of T. gondii DNA was recorded in 10 (3.6%) samples investigated, including lettuce, spinach, Chinese cabbage (pak choi), red cabbage and rape samples ( Table 1, Fig. 2). Sequencing of positive samples and comparison with the T. gondii sequence deposited in the GenBank displayed that the obtained nested PCR products were fragments of the T. gondii B1 gene (Table 2). Samples representing remaining types of vegetables were negative for Toxoplasma DNA ( Table 1). The IPC test excluded presence of PCR inhibition in investigated samples that could influence the results (details are available in Supplementary Table S3).
Regarding sampling period, the majority of vegetable samples positive for Toxoplasma DNA were detected during summer months, 5/104 (4.8%) and 2/40 (5%) in July and August respectively (Table 1). Lower number of positive samples was noted during winter and early spring season. Presence of parasite was confirmed in only 3/100 (3%) of vegetable samples collected in January and none of the samples collected in March 0/35 (0%) ( Table 1). Chi-square test, however didn't show statistical dependence between number of positive samples and month χ 2 = 2.08 (p = 0.55). Frequency of positive vegetable samples collected during particular sampling months against the total number of vegetable samples collected in this study (n = 279) was as follows: 5/279 (1.8%), 3/279 (1.1%) and 2/279 (0.7%) for July, January and August respectively ( Table 1).
Results of quantitative real-time PCR enabled the calculation of equivalent of T. gondii oocysts present in positive vegetable samples, in the final suspension obtained after flocculation ( Table 2). The approximate level of contamination of positive vegetables samples ranged between less than one and 27000 T. gondii oocysts per sample, with majority not exceeding several oocysts per sample (Table 2) (detailed information about qPCR including standard curve and positive samples is available in Supplementary Tables S1 and S2).  35 : pink refers to provinces where type II is prevalent; yellow refers to provinces where Chinese 1 genotype is prevalent and type I is common; olive represents province with scarce data about prevalence of T. gondii genotypes.

Discussion
In the present study we successfully detected Toxoplasma DNA in fresh vegetable samples collected from local open markets located in the Xining City, the capital of Qinghai province in China. To the best of our knowledge, this is the first investigation describing T. gondii detection in vegetable samples collected from territory of P.R. China using sensitive molecular tools. T. gondii DNA was detected in ten out of 279 tested samples. The results of our study confirmed that fresh vegetables available in this part of China were contaminated with T. gondii and they may pose a potential threat for public safety, particularly, people consuming them raw and unwashed. Contamination of vegetables goes hand in hand with contamination of surrounding environment, mainly soil and water, and the driving force behind it is felid activity. While there is an absence of studies regarding T. gondii presence in water, Toxoplasma has been detected in soil in different parts of China, both in urban 72,74,75 and agricultural areas 71,73,74 . These studies included Qinghai and Gansu provinces in the north-western part, Harbin and Wuhan in the centre and Nanjing in the east and all of them showed over 10% of contamination rate, in some cases 30% [71][72][73][74][75] . Most probably stray cats play a pivotal role in epidemiology of toxoplasmosis; they roam freely and may void invasive oocysts of the parasite to the environment with faeces. It is possible that these animals may reach areas where fruit and vegetables are growing and contaminate it with the Toxoplasma oocysts. In China, epidemiological surveys have been conducted regarding prevalence of T. gondii in cats; a systematic review of publications between 1995 and 2016 included 38 studies and estimated the average infection rate in 20%, and seroprevalence in stray cats was significantly higher than in pet cats 76 . Investigations performed in north-western China in Lanzhou City, Gansu Province showed 21.3% examined cats infected including 15.6% and 45.2% for household and stray cats respectively 77 . In QTP, types of Felidae have also been speculated as a potential source of T. gondii, such as Pallas' cat (Otocolobus manul) and Chinese mountain cat (Felis bieti), both species populating the plateau in high numbers 78 . Therefore, the contaminating agent and background contamination are well Presence of T. gondii DNA has been confirmed in different type of vegetable samples investigated in this study including lettuce, spinach, pak choi, rape and red cabbage samples. We have noticed differences in number of positive vegetable samples between seasons in which they were collected. Toxoplasma DNA was more frequently found in samples collected during summer months in comparison to winter and early spring. Different factors may influence this situation, such as felid activity, water or aeolian erosion and agricultural practice. Vegetables may originate from greenhouses or farmlands; when they are grown on open fields they are exposed to the rain and wind that enable transportation of soil particles and oocysts from soil surface on the vegetables. The irrigation practice may provide a similar mechanism of transportation, from soil to vegetables, or become the contamination agent in cases where the irrigational water is contaminated. Moreover, open fields are accessible to animals including felids that may excrete the oocysts of T. gondii and therefore, we have found higher number of positive samples among collected during warm months when animal activity is rising and the use of greenhouses is reduced.
The level of contamination of collected samples in this study was 3.6%, which is lower than what has been previously reported in other countries 28,29 . These studies included collection of vegetables covering a spectrum of agricultural practices, logistics and distribution as well as environmental conditions. The study conducted in Paraná, Brazil, with a tropical climate, showed 3.8% of vegetable samples collected from sales outlets and production sites positive to Toxoplasma 29 . In the same area, vegetables sold in community fairs, producers' market and generally open markets, from local producers with the corresponding logistics, and showed 9.5% samples contaminated 26 . The study which was conducted in northern Poland, an area with oceanic climate, included fruit and vegetables from open markets, from direct producers but also from greengrocers and supermarkets, while the products originated from the general area of sampling; the overall rate of contaminated samples was 9.7% 28 . On the other hand, a broad study in major Canadian cities, under completely different conditions, showed quite lower contamination level (0.26%) 27 ; it regarded imported pre-packaged or bulk green vegetables from USA and Mexico, with the corresponding logistics. In the aforementioned study in Poland, samples taken from supermarkets, where washing and sanitizing methods may have been applied, showed no contamination 28 . Our study regards, on one hand vegetables supplied in open markets, but on the other they are transported hundreds of kilometres away from the cultivation site and provided through a centralised system to the small sellers in these markets, constituting a combination of the previous examples. Therefore they may have more sophisticated packaging than in the cases of direct produce or decentralised distribution, and possibly some form of washing or sanitizing for a part of them.
The primary contamination of vegetables during cultivation has been established in a number of cases 26,28 . It should be emphasized that while comparing data from cited studies, different factors that could influence the results should be taken under consideration such as limitations of recovery, concentration and detection methods used, as well as the environmental contamination background of the investigated region, we may also assume that the handling of vegetables after their collection influences the contamination level and the threat for the public. Nevertheless, a coordinated screening of soil, vegetables and the end product at the hands of the consumer should be organised to properly assess the risk at each level of production.
Genotyping of positive samples collected in this study indicated that T. gondii isolates represent type I (eight samples) and II (two samples). Although, T. gondii -Chinese 1 genotype is widely prevalent in China with a rising gradient from west to east, in the northwest of China where Qinghai is situated, the predominating genotypes are type II and I followed by Chinese 1 35    www.nature.com/scientificreports www.nature.com/scientificreports/ -type I genotype is common 35 (Fig. 1). The study of agricultural soil contamination with Toxoplasma could help establish a hypothesis, but such studies in China have not included genotyping 71,73,74 . Hence, we assume that the localities where the particular vegetables were cultivated were contaminated with the aforementioned genotypes.
Additionally, an equivalent of the approximate oocyst charge using qPCR was established to a mean level of positive samples. The order of magnitude of the number of oocysts that might be present in positive samples ranged from one to 10 4 . For half of samples number of oocysts was below 10, and only two of them were highly contaminated. Actual contamination, however, could be even higher than demonstrated due to loss of material during the recovery process. Although the infective dose (number of oocysts) responsible for the development of infection in humans is unknown, the results indicate that transmission of the parasite to Chinese citizens by consumption of such vegetables raw and unwashed should be taken into consideration. A similar study performed in Poland showed that number of T. gondii oocysts present in vegetable samples collected from open markets and greengrocers was less than ten 28 .
Although, the morphological integrity, viability, biological activity or virulence of Toxoplasma isolates was not investigated in this study, and consequently their ability to infect humans and/or animals remains unknown, the detection of T. gondii DNA and evaluation the probable intensity of contamination of with T. gondii oocysts in vegetable samples is clear evidence for the presence of the parasite in the food chain and indicates a potential risk for humans.
Consumption of unwashed vegetables and fruits contaminated with Toxoplasma has been ranked amongst the highest risk factors for toxoplasmosis in Norway, due to the regularity these are consumed 70 . In parts of China, poor eating habits like neglect of hand washing before meal or dinner preparation has been indicated as a risk factor of toxoplasmosis infection 67 . In areas with high risks for public health, however, where traditional communal ties have not been ruptured, society, through its historical experience, has adopted countermeasures to alleviate such problems. In the case of Xining City, the general contamination background in QTP regarding parasites, which includes other dangerous parasites such as Echinococcus spp. and other Taeniidae, has forced the society to adapt by customizing its habits accordingly, like boiling water and vegetables before consumption. This is also a most probable reason why QTP and China in general, displays lower toxoplasmosis levels than the world average. Nevertheless, societal changes, like urbanization or import of different customs, like society in QTP is experiencing now, may rupture such ties, and this is when awareness and education helps in the prevention of threats that were historically managed, and general improvement of public health. Hence, protection against toxoplasmosis and other public health threats can be achieved by adoption of hygiene measures and practices in open markets, as well as sanitary regulations and best practices for larger units and centralised distribution centres, like warehouses and supermarkets.
Considering the possibility of working with contaminated vegetables and fruit as a risk factor of toxoplasmosis, a comparison with the meat industry can be performed. A study in Mexico has concluded that toxoplasmosis is not associated with handling unwashed and raw vegetables in areas with relatively low prevalence of Toxoplasma like in China 79 . Therefore, we can assume that the relatively low prevalence rate in vegetables and the occupational characteristics, regarding the handling of vegetables, do not constitute an additional risk for the workers in this sector, in contrast with other agro-industries like the meat producing. This is in contrast with the meat processing and dairy industry, where Toxoplasma has been established as a potential occupational risk factor [80][81][82] , accounting for the differences during the production processes and the conditions workers are exposed. However, more data, from different provinces of China, focused on the presence of Toxoplasma in vegetables is needed to draw unequivocal conclusions.

conclusions
Toxoplasmosis is a persistent problem in China. In this study, we report contamination of vegetable samples available in local open market in Xining City in the Qinghai-Tibet Plateau, in China, with T. gondii. The detection of T. gondii DNA in vegetable samples is evidence of the presence of the parasite in the agricultural environment and its persistence through multiple stages in the food sector, indicating a potential risk for humans. Therefore, results of our study call for more concentrated screening of the environmental matrices, soil, vegetables and the different production stage to establish the transmission routes, and the level of risk to contract toxoplasmosis through contact with contaminated vegetables. Furthermore, revision and adoption of food safety practices in order to prevent T. gondii from reaching the consumers through the food chain, in this and other parts of China.

Recovery and concentration of oocysts from vegetable samples. At first vegetables were rinsed
thoroughly. In order to concentrate and recover Toxoplasma gondii oocysts from washing suspensions, Al 2 (SO 4 ) 3 flocculation methodology was employed 83 . Finally, the suspension obtained was preserved at −20 °C for further analysis.
DnA extraction. Prior to DNA extraction, the material obtained from vegetable samples was prepared using ten freeze-thaw cycles (using liquid nitrogen and a water bath) to destroy the walls of the oocysts and improve the efficiency of DNA extraction. Afterwards, DNA extraction was performed using a commercial TIANamp Micro DNA Kit (DP 316) (Tiangen Biotech, Beijing, China) according to the manufacturer's instructions. The DNA was then stored at −20 °C. www.nature.com/scientificreports www.nature.com/scientificreports/ (A&A Biotechnology, Gdynia, Poland), 400 nM of each primer (Genewiz SZ, Suzhou, China), 80 nM of TaqMan probe (Genewiz SZ, Suzhou, China), and 5 μL of template DNA in a 25 μL reaction volume. Amplification was performed with an initial polymerase activation step (10 min at 95 °C), followed by 40 cycles of denaturation (15 s at 95 °C) and hybridisation/extension (1 min at 60 °C) in a AriaMx Real-time PCR thermocycler, Agilent). PCR products were analysed using AriaMx Real-time PCR System Software. The cycle threshold (CT) value, determining the cycle number at which the reporter's fluorescence exceeds the threshold value, was recorded. A sample was considered positive if the CT value was < 40.

Specific detection of T. gondii
All PCR experiments were performed including T. gondii positive controls to ensure the correct functionality of the reaction, as well as negative controls to ensure that no PCR component had been contaminated. The DNA isolated from tachyzoites of parasite (the Toxoplasma gondii RH strain), obtained from the National Institute of Hygiene, Poland was used as positive control in the performed experiments.
Additionally, samples were retested for the presence of PCR inhibitors by mixing 4 μL of DNA template and 1 μL of internal positive control (IPC) (more details concerning the IPC used are available in Supplementary information). Comparison of results of amplification (Cq values) obtained for samples containing combination of IPC and template DNA with IPC alone allowed estimation of potential interference.
Sequencing. The final PCR products from the positive samples were sequenced. Both forward and reverse orientation cycle sequencing was performed using the amplification primers. The sequences obtained were then analysed using GeneStudio TM Professional (GeneStudio, Inc., USA) and ChromasDNA sequencing software.
Amplification of genetic markers using nested PCR. The 1 st step of nested PCR was performed using set of external primers (Table 3) in a 25 μL reaction volume and the amplification reaction mixture consisted of 12.5 μL of the standard and ready-to-use PCR mixture 2xPCR Mix Plus High GC (A&A Biotechnology, Poland) containing recombinant Taq polymerase, PCR buffer, magnesium chloride, nucleotides, stabilisers, and gel loading  www.nature.com/scientificreports www.nature.com/scientificreports/ buffer, 200 nM of each forward and reverse primer (Genewiz SZ, Suzhou, China), and 2 μL of template DNA. Amplifications were performed with an initial polymerase activation step (5 min at 95 °C), followed by 35 cycles of denaturation (30 s at 94 °C), primers annealing (1 min at 55 °C), strand extension (2 min s at 72 °C), and final extension (7 min at 72 °C). The 2 nd step of nested PCR reactions were performed using internal set of primers (Table 3) under following amplification reaction mixture conditions: 12.5 μL of the standard and ready-to-use PCR mixture 2xPCR Mix Plus High GC (A&A Biotechnology, Poland), 400 nM of each primer (Genewiz SZ, Suzhou, China), and 2 μL of template DNA in a 25-μL reaction volume. Amplifications were performed according to the same protocol as in the first reaction, with the exception that annealing temperature was 60 °C.
Restriction analysis of nested PCR products. In order to perform restriction analysis 10 µL of PCR product were mixed with 2 µL of 10x digestive buffer and 1 U of restriction enzyme (each of the two restriction enzymes in case of double digestion) (Thermo Scientific, USA). The reaction was carried out according to manufacturer's instruction. RFLP products were analysed using a WD-9413B gel imaging analysis system (Beijing Liuyi Biotechnology, China) following electrophoresis on a 3% gel agarose (Biowest Regular Agarose G-10, Gene Company) stained with ExRed nucleic acid electrophoresis dye (Beijing Zoman Biotechnology, China).
Determination of initial copy numbers of T. gondii DnA using qpcR. In order to determine the initial copy number of the detected T. gondii DNA, each positive sample was tested with qPCR based on a standard curve. The standard template was generated by cloning the insert gene (a fragment of the B1 gene amplified by PCR using the set of primers described above) into a plasmid. To generate the standard curve, three series of nine dilutions of standard DNA in the range from one to 10 8 DNA copies per one μL were prepared (Fig. 3). Following amplification of the standard dilution series, the standard curve was obtained by plotting the log of the initial template copy number against the CT value generated from each dilution. Amplification of the standard dilution series and of the DNA isolated from vegetable samples was run on the same plate. Comparing the CT values of the unknown samples with the standard curve thus enabled the quantification of initial copy numbers.
Initial copy numbers of T. gondii DNA were used to calculate equivalent of T. gondii dispersive forms that might be present in the examined samples after flocculation (recovery procedure). According to our previous experiments, using real-time PCR, it is possible to detect single T. gondii oocyst in water suspension 13 . Thus, the number of T. gondii oocysts (oocyst charge of the sample) was estimated using following formula: where: OCS is the oocyst charge of the sample; N is the initial copy number determined using qPCR (per amount of DNA template taken for PCR); 35 is the number of copies of the B1 gene per one T. gondii cell; A = 8, the number of T. gondii sporozoites per one oocyst; B is the multiplication factor referring to a total amount of DNA extracted from investigated sample.

Statistical analysis.
In order to investigate the correlation between T. gondii presence in investigated vegetable samples and sampling period statistical analysis was performed using the Pearson chi-square test (χ 2 ). Calculations were performed using MATLAB 2015a.

Data availability
All data generated or analysed during this study are included in this published article.