Direct detection and genotyping of Toxoplasma gondii in meat samples using magnetic capture and PCR

https://doi.org/10.1016/j.ijfoodmicro.2010.02.027Get rights and content

Abstract

Different transmission routes, including the ingestion of undercooked meat, can result in Toxoplasma gondii infection in humans. The development of effective prevention strategies is hampered by a lack of quantitative information on the contamination level of different types of meat. Therefore, we developed a method for detection and quantification of T. gondii. The method involved preparation of crude DNA extract from hundred gram samples of meat, magnetic capture of T. gondii DNA and, quantitative real-time PCR targeting the T. gondii 529-bp repeat element. The detection limit of this assay was approximately 230 tachyzoites per 100 g of meat sample. There was a linear relation between the number of parasites added to the samples and Cp-values. Results obtained with the PCR method were comparable to bioassay results for experimentally infected pigs, and to serological findings for sheep. In addition, the T. gondii in 50% of the positive sheep samples could be genotyped by sequencing of the GRA6 gene, after isolation of the gene by magnetic capture. Two subtypes of GRA6 type II were identified in the 16 samples from sheep. For seven samples, the identification of T. gondii as type II was confirmed by microsatellite typing. The PCR method can be used as an alternative to bioassay for detection and genotyping of T. gondii, and to quantify the organism in meat samples of various sources.

Introduction

Infections by the protozoan parasite Toxoplasma gondii are widespread in humans and in many warm-blooded animal species. Although most infections in humans are asymptomatic, severe complications may occur in immunocompromised patients and after congenital T. gondii infection (Weiss and Dubey, 2009). The incidence of congenital toxoplasmosis in the Netherlands was recently estimated at 2 per 1000 live-born children (Kortbeek et al., 2009). Using these new data the disease burden was calculated at 2300 disability-adjusted life years (Kortbeek et al., 2009), which is almost four times higher than the previous estimate (Havelaar et al., 2007). This high disease burden makes toxoplasmosis one of the most important food-borne infections, and underscores the necessity to intervene. Considering the lack of evidence of the effectiveness of treatment (Thiebaut et al., 2007, Gilbert, 2009), prevention strategies are considered most effective. Humans can contract T. gondii via tissue cysts in undercooked meat and via accidental ingestion of oocysts by contact with cat faeces, contaminated soil, water, or vegetables. Even though the sources of T. gondii infection for humans are well known, insight in their relative contribution needs to be improved for the development of effective prevention strategies. Risk-factor analysis indicates that 30 to 63% of human infections can be attributed to the consumption of undercooked meat (Cook et al., 2000). However, what kind of meat contributes most to human infections depends on prevalence of T. gondii in consumption animals and on eating habits. An indication of the relative contribution of different kinds of meat can be obtained by screening large numbers of meat samples for the presence of T. gondii. Genotyping isolated parasites will give further insight into the epidemiology of toxoplasmosis.

The gold standard for detecting T. gondii in meat samples is a bioassay using either mice or cats. These bioassays are laborious and time-consuming techniques, which are not desirable for screening large numbers of samples from an animal ethics point of view. Therefore, PCR-based methods to detect T. gondii in meat samples have been developed. However, although the PCR itself is usually sensitive in detecting T. gondii DNA, when used on meat samples, these methods lack sensitivity in comparison to the bioassay (da Silva and Langoni, 2001, Garcia et al., 2006, Hill et al., 2006). This lack of sensitivity of PCR-based methods is likely due to the inhomogeneous distribution of T. gondii tissue cysts, in combination with the small size of the sample: For PCR, DNA is usually isolated from 50 mg of sample at maximum. In the bioassay either up to 500 g of meat is fed to a cat, or fifty to a hundred grams of meat is inoculated into mice after artificial digestion. Clearly, the probability of the presence of a tissue cyst in a 50 mg sample is much lower than in the 50 to 500 g sample used in bioassays. As a consequence, taking fifty milligrams of the homogenate of a large sample, instead of taking a fifty milligram sample randomly, will increase the probability of isolating T. gondii DNA. However, it will be present at a low concentration in a high background of host DNA, which might lead to inhibition of the PCR (Bellete et al., 2003). The effects of a low concentration and inhibition can be overcome by sequence-specific magnetic capture, as has been previously described for the detection of mycobacterial DNA in clinical samples (Mangiapan et al., 1996). To detect T. gondii, the 529-bp repeat element (Homan et al., 2000, Reischl et al., 2003) is used as target for sequence-specific capture and real-time PCR. Because the T. gondii genome is distributed over fourteen chromosomes (Khan et al., 2005), only part of the genome is isolated using sequence-specific capture. The 529-bp repeat element is highly conserved (Reischl et al., 2003), and therefore not suitable for typing. To enable genotyping, the GRA6 gene is captured. GRA6 is a dense granule protein of 32 kDa (Lecordier et al., 1995). The single copy GRA6 gene is highly polymorphic (Fazaeli et al., 2000), which makes it a useful marker for typing.

It was our aim to develop a PCR-based method that can be used as an alternative to the bioassay in quantitative screening of large numbers of meat samples. In this paper, a method that combines homogenization of a large sample with sequence-specific magnetic capture to detect and genotype T. gondii in meat samples is described. This method simplifies testing large numbers of meat samples, to determine the relative contribution of different kinds of meat in human T. gondii infections, while reducing the use of experimental animals.

Section snippets

Oligonucleotides

All Tox-oligonucleotides are complementary to the 529-bp repeat element (GenBank AF146527) (Homan et al., 2000), and all GRA6-oligonucleotides are complementary to the GRA6 gene (GenBank L33814) (Lecordier et al., 1995) (Table 1). All oligonucleotides were synthesized by Biolegio (Nijmegen, The Netherlands). Capture-oligonucleotides were designed to capture either both strands of the 529-bp repeat element (Tox-CapF and Tox-CapR), or both strands of the GRA6 gene (GRA6-CapF and GRA6-CapR), from

Detection limit of the magnetic capture-PCR and comparison of isolation methods

The 95% detection limit of the PCR was estimated at 15.7 fg (95% CI: 10.0–55.9 fg) per PCR reaction. The probit regression model adequately fitted the data (Pearson's χ2 (df = 13) is 10.752, p = 0.632). Using probit analysis on the results with spiked meat samples, the 95% detection limit of the method including the magnetic capture was estimated at 227 tachyzoites per 100 g sample (95% CI: 107–3094). Model fit was adequate (Pearson's χ2 (df = 7) is 10.152, p = 0.180).

Seventy-three sheep and 28 pig

Discussion

This paper describes a PCR-based assay for detecting and genotyping T. gondii tissue cysts in hundred gram meat samples using sequence-specific magnetic capture of T. gondii DNA followed by real-time PCR targeting the 529-bp repeat element.

To develop this method, first a sensitive real-time PCR for the 529-bp repeat element (Reischl et al., 2003) was adapted to the available PCR-system and a competitive internal amplification control was added. The detection limit of 16 fg corresponds to 0.14

Acknowledgements

This work was financially supported by the Dutch Food and Consumer Safety Authority (VWA).

We thank Lothar Züchner (VWA, Zutphen, The Netherlands), and the personnel at the sheep slaughterhouses in Dodewaard and Twello for their help with collecting the field samples. The assistance of Jeroen Roelfsema, Sietze Brandes and Jan Cornelissen was highly appreciated.

References (31)

  • A. Kijlstra et al.

    The role of rodents and shrews in the transmission of Toxoplasma gondii to pigs

    Veterinary Parasitology

    (2008)
  • L. Lecordier et al.

    Characterization of a dense granule antigen of Toxoplasma gondii (GRA6) associated to the network of the parasitophorous vacuole

    Molecular and Biochemical Parasitology

    (1995)
  • C. Richomme et al.

    Genetic characterization of Toxoplasma gondii from wild boar (Sus scrofa) in France

    Veterinary Parasitology

    (2009)
  • L.M. Weiss et al.

    Toxoplasmosis: a history of clinical observations

    International Journal for Parasitology

    (2009)
  • D. Ajzenberg et al.

    Genotype of 88 Toxoplasma gondii isolates associated with toxoplasmosis in immunocompromised patients and correlation with clinical findings

    Journal of Infectious Diseases

    (2009)
  • Cited by (0)

    View full text