OUP user menu

How to detect Toxoplasma gondii oocysts in environmental samples?

Aurélien Dumètre , Marie-Laure Dardé
DOI: http://dx.doi.org/10.1016/S0168-6445(03)00071-8 651-661 First published online: 1 December 2003


Detection of Toxoplasma gondii oocysts in environmental samples is a great challenge for researchers as this coccidian parasite can be responsible for severe infections in humans and in animals via ingestion of a single oocyst from contaminated water, soil, fruits or vegetables. Despite field investigations, oocysts have been rarely recovered from the environment due to the lack of sensitive methods. Immunomagnetic separation, fluorescence-activated cell sorting, and polymerase chain reaction have recently shown promising use in detection of protozoa from complex matrices. Such procedures could be applied to T. gondii detection, if studies on the antigenic and biochemical composition of the oocyst wall are completed. Using such methods, it will be possible to assess the occurrence, prevalence, viability and virulence of T. gondii oocysts in environmental matrices and specify sources of human and animal contamination.

  • Toxoplasma gondii
  • Oocyst
  • Felid
  • Water
  • Environment
  • Detection

1 Introduction

Toxoplasmosis is a widely prevalent zoonosis in humans and warm-blooded animals world-wide, due to the tissue cyst-forming coccidium, Toxoplasma gondii. Toxoplasmosis is generally asymptomatic except after congenital transmission, when associated with abortions or clinical diseases, and in immunocompromised patients. Cases of severe infection are sometimes recorded in immunocompetent people [1]. Transmission occurs by consumption of undercooked or raw meat containing tissue cysts or by ingestion of resistant oocysts from environmental matrices (soil, water, fruits and vegetables). The importance of each route in toxoplasmosis transmission is undetermined, since it is not yet possible using serological investigations to discriminate infections due to oocysts from those induced by cysts [2]. However, the impact of oocysts in toxoplasmosis epidemiology needs to be specified because (i) they are suspected to be associated with high T. gondii seroprevalence in some communities [3], (ii) they are recognised as the source of large emerging outbreaks of acute toxoplasmosis in humans from soil or water [49], (iii) they are probably responsible for a significant part of infections in animals that could be later consumed by humans [10].

Although T. gondii oocysts persist a long time in the environment, some of their features may hamper their detection. They are shed exclusively by felids, in contrast to other waterborne protozoa such as Cryptosporidium parvum and Giardia intestinalis, often in large quantities (up to 108) but during a brief period (1–3 weeks) and then they are disseminated by biotic and abiotic factors [11]. Thus, the number of oocysts which may be found in random environmental samples is probably low. In addition, they may be confused with Hammondia sp. and Neospora sp. oocysts, two closely related coccidia that may occur in the environment [12, 13]. Debris in samples further complicates oocyst detection. Microscopy and bioassays in mice are often unsuited to sensitive and simple detection. Therefore, development of new detection methods is necessary. Specific and sensitive methods exist for other protozoa [14] but they have not yet been developed for T. gondii oocysts. In this paper, we review oocyst dissemination in the environment and propose methods for their detection, especially in water, based on the analysis of methods used for other protozoa.

2 T. gondii life cycle in definitive hosts

Felids (domestic and wild felids) are the only known definitive hosts of T. gondii [11]. In cats (Felis catus), natural contamination occurs soon after weaning through ingesting tissue cysts in infected preys (small mammals and birds). Vertical transmission is rare and experimental; contamination by tachyzoites or oocysts leads to irregular oocystogenesis [15, 16]. Based on serologic surveys, up to 74% of the adult cat population is infected by T. gondii [10]. Seroprevalence increases with age and is higher in free-roaming cats, which hunt for food, than in domestic cats, which are often fed with preserved food [17]. In natural conditions, cyst-induced contamination is probably responsible for most infections in wild felids (including cougars, jaguarundis, mountain lions, and bobcats) [18] but experimental data have shown that oocyst production is more variable than in domestic cats [19, 20].

In cats, prepatent periods to oocyst shedding vary with the parasite stage ingested: 3–10 days after ingesting bradyzoites in tissue cysts and 19 days or more after ingesting tachyzoites or oocysts [15, 16, 21]. The sexual cycle has been well documented only after cyst contamination [21]. In a cat's gut, released bradyzoites invade enteroepithelial cells to undergo asexual multiplication. Sexual differentiation into female macrogametocytes or into male microgametocytes begins 3–15 days after contamination. Flagellate microgametes are released from microgametocytes into the intestinal lumen to fertilise macrogametes in enteroepithelial cells, but fertilisation seems not to be essential for successful T. gondii reproduction [22]. Immature oocysts (10×12 μm) are released from epithelial cells and passed daily in faeces. They are not directly infective for animals and humans. Development of oocyst infectivity (i.e. sporulation) occurs in 1–21 days under definite temperatures (11–25°C) and favourable aerobic and humidity conditions [23]. One sporulated oocyst (11×13 μm) has two sporocysts (6×8 μm), each containing four infective sporozoites (2×6–8 μm). Sporozoites are similar ultrastructurally to bradyzoites in tissue cysts and tachyzoites [24].

Short shedding time (1–3 weeks) is balanced by production and dissemination of large quantities of environmentally resistant oocysts (107–108 for one cat). It was reported that the proportion of domestic cats excreting oocysts at any one time is 2% in most countries [11]. Re-shedding was demonstrated experimentally after a second challenge with T. gondii, after corticoid treatment or superinfection by Isospora felis [2527].

3 Resistance of infective oocysts

3.1 Environmental conditions

Oocysts are spread in the environment mainly via wind, water, manure and by earthworms and arthropods: they can contaminate surface water, soil, harvest feeds, fruits and vegetables [11, 2831]. They have been recovered from naturally contaminated soil but never from water [11].

Unsporulated oocysts lose their capacity to sporulate after freezing (1 day at −21°C or 7 days at −6°C) and heating (50°C, 10 min) [23, 32]. Refrigerator conditions (4°C for 6–11 weeks) are not sufficient to prevent development of oocyst infectivity [33]. As shown in Table 1, sporulated oocysts can remain infective in water for at least 54 months at 4°C [35] and in experimentally infected soils for 18 months under various temperatures [29]. Increasing temperatures results in decreased survival time. Freezing may not be sufficient to kill sporulated oocysts as they can survive for 28 days after constant freezing at −21°C [32]. In addition, fluctuating temperatures do not notably affect oocyst survival time both in water and in soil and oocysts can remain infective after severe winters [29, 32]. Drying of uncovered suspensions or by direct exposition to sunlight is moderately deleterious to sporulated oocysts [11, 34, 36]. Oocysts probably survive in marine waters since T. gondii infections have been reported in marine mammals [11]. Oocysts can be removed by oysters from experimentally contaminated sea water [39]. Thus, molluscs can act as a potential source of T. gondii infections due to their consumption by humans or marine mammals, as shown for C. parvum oocysts [40, 41]. Cool and moist fruits or vegetables may provide an optimal environment for oocyst survival. Kniel et al. have recently shown an 8-week survival time of infective T. gondii oocysts on raspberries stored at 4°C [37]. These fruits have been previously reported as an important source of human infection caused by Cyclospora cayetanensis, a closely related coccidia to T. gondii [42].

View this table:

Duration of infectivity of T. gondii oocysts under environmental conditions

Temperature (°C)ConditionsSurvival timea,bReferences
−20In water14–28 d[32, 34]
−10/−5In water106 d[35]
0In water13 mo[35]
+4In faecal suspensions183–410 dc[36]
In faecal deposits214–410 dc[36]
In water54 mo[34]
On berries56 d[37]
+10/+25In water200 d[35]
+22.5In water306–410 dc[36]
In faecal suspensions153–410 dc[36]
In faecal deposits107–306 dc[36]
+23–29In moist potted soil117 d[28]
+30In water107 d[35]
+35In water32 d[35]
+37In water91–306 dc[36]
In faecal suspensions46–199 dc[36]
In faecal deposits30–153 dc[36]
+40In water9 d[35]
+45In water1d[35]
+50In water30–60 min[28, 35]
+55/+58In water<15 min[34, 35]
+60/+70In water<1 min[35]
−20–+35In faecal deposits in soil18 mo[29]
−6–+39In water122–306 d/153–410 dd[36]
In faecal suspensions76–306 d/91–306 dd[36]
In faecal deposits46–183 d/76–334 dd[36]
+15–+30In faecal deposits in soil56–357 d[29]
+20–+27In moist soil106 d[38]
  • aOverall infectivity of the oocyst suspension determined by bioassays in mice.

  • bAbbreviations: d, days; mo, months; min, minutes.

  • cFirst number refers to the duration of infective oocysts in uncovered suspensions; second number refers to the duration of infective oocysts in covered suspensions.

  • dUncovered (first number) and covered (second number) suspensions exposed to direct sunlight/or placed in a shady situation.

3.2 Disinfecting agents

Oocysts exhibit a remarkable resistance to various inactivation procedures and especially to chemical reagents [11, 23, 28, 34, 43, 44]. For example, oocysts remain viable in an aqueous 2% sulfuric acid or 2.5% potassium dichromate for several years at 4°C. Moreover, oocysts are resistant to detergents or disinfectant solutions such as sodium hypochlorite solutions. A 1-log diminution of infectivity (based on the oral dose of oocysts able to infect 50% of mice, i.e. ID-50) is only observed after 1–2 h contact between oocysts and 6% sodium hypochlorite solution, and a 2-log diminution only after 4 h [28]. Resistance to chemical reagents is not an isolated feature among coccidia since another waterborne organism, Cryptosporidium sp., is resistant to disinfection procedures such as chloramination during drinking water treatment [45]. Due to their resistance to chemical and physical agents, infective T. gondii oocysts could be present in water or on food. Therefore, it is necessary to develop procedures for oocyst inactivation or elimination in contaminated matrices before they become human or animal supplies. Gamma radiations could be used routinely for oocyst inactivation on fruits and vegetables [43, 44]. Oocyst resistance to disinfection procedures and to disinfectant levels found in drinking water treatment plants remains to be evaluated, but is highly probable.

4 Oocyst-transmitted T. gondii infection in intermediate hosts

4.1 In animals

Cats are essential for the maintenance of T. gondii in the environment since infections are virtually absent from areas lacking cats [46, 47]. A single oocyst is sufficient to induce infection in pigs [48] or in mice [49], but pathogenicity may depend on the strain, dose and inoculation route [50]. Oocysts are less infectious and pathogenic for cats than for intermediate hosts [15, 20]. Ingesting oocysts in water, soil or feed is probably the most common route for T. gondii infection in non-carnivorous mammals and birds. Meat-producing animals can show a very high seroprevalence throughout the world (up to 100%) [10]. However, the risk of T. gondii infection is higher in animals raised in extensive management than in intensive management. Sheep and goats on pasture show high seroprevalence in many countries (up to 92 and 75%, respectively). Most T. gondii infections in pigs are due to oocysts via consumption of contaminated feed and soil [51]. Resulting infections are of veterinary and economic importance since, for example, T. gondii is responsible for about 12% of abortions in sheep flocks, and natural oocyst-induced outbreaks of lethal toxoplasmosis were recorded in pig farms [11]. Independently of seroprevalence, prevalence of T. gondii in tissues of meat-producing animals is higher in sheep, goats and pigs than in poultry, horses and cattle [10]. In turn, humans may be infected by consumption of contaminated meat.

4.2 In humans

In humans, the percentage of infections due to oocysts remains undetermined. Owning cats may not be associated with toxoplasmosis [3, 52], demonstrating the key role of environmental matrices as sources of infection. T. gondii infection seems strongly associated with soil contact [53], contributing up to 17% of infections among pregnant women in Europe [52]. In Brazil, consumption of untreated water was significantly associated with T. gondii seropositivity with a 1.6- to 3-fold increased risk of infection [3]. Ingesting oocysts probably causes infections in some populations of vegetarians [1, 54]. To date, only a few outbreaks of acute toxoplasmosis epidemiologically linked to oocysts have been reported in overall populations (Table 2). All are linked to the presence of infected felids in areas where cases were recorded and where favourable conditions enhanced duration of infective oocysts. Outbreaks were associated with ingestion of oocysts from soil [4, 5, 7] or from water [6, 8, 9]. To date, the best-documented outbreak occurred in Victoria, Canada, in 1995 [8]. Epidemiological data suggested that infections were due to oocysts shed by domestic cats and/or cougars into a reservoir feeding the municipal drinking water system [55, 56]. Water temperature and the disinfection procedure (chloramination without filtration) enhanced survival of oocysts in the reservoir [57]. Recently, 290 people may have been contaminated by T. gondii during a waterborne outbreak in Brazil, although detailed investigations were not conducted [9]. This could establish a new record of cases for oocyst-induced acute toxoplasmosis in humans. As shown in Table 2, waterborne outbreaks are larger than those from solid matrices (soil, fruits and vegetables). During or after these outbreaks, direct evidence for soil or water contamination has never been proven, despite field investigations, except in one case [7]. This emphasises that current detection methods are not well adapted to oocyst detection in environmental samples and this highlights the necessity to develop specific and sensitive methods.

View this table:

Oocyst-transmitted outbreaks of toxoplasmosis in humans

CountryYearSuspected sourceNumber of casesOocyst isolationReference
  • aNS, not specified.

  • bSuspected outbreak, without investigation.

5 Detection in environmental samples

5.1 Sampling strategy

Proximity of cats to human homes and smaller space for deposition of cat faeces in urban areas may increase the possibility of contamination with oocysts [3]. But in rural areas, oocysts may be shed far from homes. It may be difficult to find faeces because they are quickly disintegrated under natural conditions (about 1 week) and felids often bury them [29, 56]. In soil, oocysts persist at the surface or up to a 10 cm depth [29] but their transport through soil is unknown. In water, they may be found more likely at the bottom due to their specific gravity (1.104–1.140) [23]. Fruit and vegetable contamination likely occurs by direct exposure to cat faeces or irrigation with contaminated water. Due to oocyst persistence and to the possibility for cats to be infected at any season during the year, no seasonal occurrence of oocysts is expected in most of cases. However, environmental matrices may be more heavily contaminated by soil washing after peaks in rainfall [8, 58]. Association with clusters of persons acutely infected and peaks in rainfall and turbidity was shown during the Victoria's outbreak [8]. If contamination sources are not identified, oocysts should be searched for where humans and animals can be contaminated due to the proximity of cats, and where favourable conditions allow for oocyst persistence, especially in moist and shady areas. Locations for human contamination are likely to include gardens, public enclosures, backyards, drinking water systems (from reservoirs to faucets), and lakes and pools for recreational activities, which can be contaminated directly or indirectly (runoff) by cat faeces. For animals, epizootic (outbreaks) or enzootic situations can provide information on contaminated sources such as meadows around farms, watering places, ponds, streams and harvest feeds where cats can deposit faeces. Oocyst detection from marine water may be problematic due to their low density in this matrix, even if marine molluscs can concentrate oocysts in their tissues [39]. For overall purposes, sampling should be done as soon as possible after the first recorded cases. This may be difficult because clinical toxoplasmosis may occur two to three weeks after infection and during this delay oocysts may disappear from the source(s) of contamination [4, 5, 57]. Due to probable heterogeneous repartition of oocysts in soil or in water, sample sizes should be as large as possible to detect low numbers of oocysts: at least 50–100 l for various turbid water samples and 100 g for solid matrices (soil, fruits, vegetables), depending on the available sample size.

5.2 Detection methods in water

5.2.1 Concentration techniques

Few methods have been described to detect T. gondii oocysts in naturally contaminated waters [51, 57]. They were unsuccessful, in part because they were poorly specific for oocysts. Detection methods of waterborne protozoa are often developed under laboratory conditions, with tap water and high inoculum size [14]. These conditions are far from those that may occur in most natural samples (turbidity, low number of target organisms). Therefore, experimental tests under various turbidity conditions with low inoculum sizes are necessary to validate techniques. Based on available methods for detection of waterborne protozoa, T. gondii recovery procedures may include the following steps: concentration of 50–1000 l by flocculation or by filtration, purification, and detection of oocysts [14]. Recently, three experimental procedures, two chemical flocculations and one centrifugation procedure, were evaluated by Kourenti et al. for the detection of T. gondii oocysts [59]. They have shown recovery rates of >80%, but infectivity for mice of recovered oocysts was evaluated only for the centrifugation procedure. However, these techniques were done with non-turbid water (demineralised water) using a small volume (1 l). Flocculation is easy to perform, inexpensive and largely used in water plant treatment. But, filtration is more robust than chemical flocculation for turbid water processing and for field investigations. Filter pore sizes (1.0–8.0 μm) used for Giardia and Cyclospora (oo)cysts recovery may catch T. gondii oocysts with minimal clogging or could be combined depending on sample turbidity, since their size (10–13 μm) is close to the size of Cyclospora oocysts (8–10 μm) and to those of Giardia cysts (7–12 μm) [60, 61]. A pre-filtration step with successive decreasing mesh sieves could be incorporated to remove large particles.

For other waterborne protozoa, large volume processing can be achieved through polypropylene or cotton cartridges filters but oocyst elution may be time-consuming to improve recovery [62]. Frontal filtration with cellulose acetate membranes (CAM) or polycarbonate membranes is useful for small volumes (up to 50–100 l) both in the field and the laboratory. They can give better recovery rates than cartridges filters from turbid waters due to an easier elution procedure (49% versus 12%, respectively) [63]. More recently, safe-handle sampling capsule filters, such as Gelman Envirochek standard (GES) filter (designed for 10 l), Gelman Envirochek high volume (GEHV) filter (designed for >500 l) and the IDEXX Filta-Max system were developed for use under various conditions. GES and GEHV have shown a 50% overall recovery of Giardia cysts in low turbid waters [64, 65]. However, when tested with seeded high turbid waters, filtration capacities and recoveries decreased (<1% recovery at 99 nephelometric turbidity units (NTU)) [65]. With the IDEXX Filta-Max system, up to 70% of Giardia cysts are recovered both from 50 l of seeded tap water (<0.05 NTU) or seeded matrix samples (1.0 to 15.6 NTU) [66]. These systems and others were introduced as an improvement in standard procedures for protozoa monitoring in water supplies [67, 68]. Despite their cost, they could be used for T. gondii detection according to water turbidity. C. cayetanensis oocysts were successfully recovered from wastewater using GES [69]. Detergents improve recovery rates by breaking hydrophobic interactions between protozoa and filter components during elution [70]. Basically, elution buffers contain 0.01–1% detergents such as laureth-12, polyoxyethylensorbitan mono-oleate (Tween-80) and/or sodium dodecyl sulfate (SDS) [14]. Since T. gondii oocysts remain infective in 1% Tween-80 or 1% SDS in water for 24 h [28], they may be eluted with usual elution buffers. CAM dissolution in 100% acetone is an alternative to the elution procedure [71]. However, it seems not to be appropriate for virulence studies because viability of protozoa is reduced after dissolution [72].

5.2.2 Purification techniques

Following elution, oocysts may be concentrated in a pellet by centrifuging the eluate. They could then be purified by various techniques discussed in this paragraph. Sucrose flotation (specific gravity ≥1.15) is commonly used to purify T. gondii oocysts (1.104–1.140 specific gravity) from complex matrices [11, 23]. But flotation may not be sufficient to remove debris with the same specific gravity and substances that inhibit molecular detection. Therefore, flotation is not appropriate for detection of particularly low numbers of oocysts. In addition, gradients do not distinguish T. gondii oocysts from Hammondia sp. and Neospora sp. oocysts, two closely related coccidia that may occur in environmental samples [12, 13]. Immunomagnetic separation (IMS) and fluorescence-activated cell sorting (FACS) may be substituted for flotation because they result in cleaner preparations. Depending on IMS kits, reproducible and good recovery rates (up to 100%) were reported from environmental waters seeded with Cryptosporidium and Giardia (oo)cysts [73, 74]. Use of less than 8-month-old, viable or heat-inactivated organisms did not affect IMS performances from turbid waters [75]. Since T. gondii oocyst populations are probably heterogeneous in age and viability state, IMS with appropriate monoclonal antibodies (mAbs) could recover all oocysts. FACS was performed to sort and/or to count Cryptosporidium, Giardia and, recently, C. cayetanensis (oo)cysts suspensions [14, 76]. However, background fluorescence resulting from autofluorescence of debris, or non-specific mAbs binding may hamper sorting. Autofluorescence of T. gondii oocysts could be used in a FACS assay as shown for C. cayetanensis [76].

To date, no mAbs are available against the T. gondii oocyst wall. A similar situation occurs for Cyclospora sp., Hammondia sp. and Neospora sp. oocysts. The T. gondii oocyst wall has three to five layers but their biochemistry remains to be specified [24, 77, 78]. Studies on oocyst wall antigens are rare due to the inability to obtain oocysts in vitro and difficulties to have clean preparations of oocyst walls. Kasper et al. only produced mAbs against the inner layer of the oocyst wall [79]. They assumed that the outer layer may be non-antigenic. Ferguson et al. described a protein recognised by an anti-MIC4 mAb (i.e. directed against a microneme antigen) but it seems not to be associated with the outer layer [80]. Interest in mAbs could be revived since Everson et al. have recently described a method for isolation of oocyst walls on iodixanol gradient [81]. They could serve as starting material for mAb production, as previously described [82, 83]. Due to nearly identical antigenic profiles between T. gondii and H. hammondi [84, 85], similarities between oocyst wall antigens could occur, necessitating a specific screening of chosen mAbs against other coccidia.

5.2.3 Microscopic detection

Light microscopy may be sufficient to detect oocysts in highly contaminated samples (i.e. cat faeces). Examination under an ultraviolet beam (excitation filter: 330–385 nm, dichroic mirror: 400 nm, barrier filter: 420 nm) facilitates detection because both unsporulated and sporulated oocysts exhibit a typical blue autofluorescence similarly to other coccidia like Neospora, Hammondia, and Cyclospora species [86]. However, microscopy is often time-consuming and poorly sensitive if sample purification is insufficient and weakly specific due to risk of confusion with morphologically related coccidia, Hammondia sp. and Neospora sp. Besides, all oocysts in the same suspension do not exhibit autofluorescence under ultraviolet excitation, leading to false negatives in case of low numbers of oocysts. An IMS-immunofluorescence assay procedure could improve microscopic detection by using appropriate mAbs against cellular components of oocysts.

5.2.4 Molecular analysis

The molecular approach is attractive since detection is specific, sensitive and quick and could provide information on genotypes of isolated oocysts. T. gondii is often detected in body fluids and tissues by amplification of the B1 repeat gene by polymerase chain reaction (PCR) [87]. PCR-based detection of oocysts is less easy because (i) DNA extraction with current extraction methods is difficult due to the robustness of oocyst wall and (ii) compounds in environmental matrices may inhibit PCR, resulting in up to 100–1000-fold reductions in sensitivity [88]. Two approaches may be used to release DNA from oocysts. One is an in vitro excystation followed by sporozoite DNA extraction. A second is to destroy oocysts by different methods before DNA extraction. Some of these procedures developed for other closely related coccidia are reported in Table 3. They include oocyst disruption by freeze/thaw events or by grinding with glass beads and lengthy proteinase K digestion. However, there is no standard protocol to define the number of freeze/thaw cycles to obtain sufficient disruption and glass bead procedures could be inefficient for disruption of low numbers of oocysts [91]. Lysis with saline saturated buffers including proteinase K and cetyl trimethyl ammonium bromide (CTAB) have shown promising results for related coccidia [91].

View this table:

DNA extraction protocols for T. gondii-related coccidia

CoccidiaMatrixExtraction procedureDetection limit (number of oocysts)Reference
Cyclospora cayetanensisWastewaterZinc sulfate purified oocysts10[69]
Resuspension in a 1× PCR buffer (50 mM KCl, 10 mM Tris–HCl (pH 9.0), 0.1% Triton X-100, 1.5 mM MgCl2) and 6% resin matrix
Seven freeze–thaw cycles (dry-ice ethanol/water bath at 98°C)
Surpernatant directly used for PCR
RaspberriesGlass wool column purification10–30[89]
Extraction-using FTA filters (Fitzco, Inc.)
Eimeria sp.FaecesNaClO and Percoll (Amersham) purified oocysts250[90]
Lysis in 10 mM Tris–HCl, 10 mM EDTA, 0.5% SDS, pH 8.0 and proteinase K (20 mg ml−1)
Vortexing with 2 mm glass beads
Extraction with QIAamp DNA Mini Kit (Qiagen)
FaecesNaClO and Sucrose purified oocysts10[91]
Lysis in 660 mM EDTA, 1.3%N-lauroylsarcosine, pH 9.5 and proteinase K (2 mg ml−1) followed by incubation with a CTAB-lysis buffer (2% (w/v) CTAB, 1.4 M NaCl, 0.2%β-mercaptoethanol, 20 mM EDTA, 100 mM Tris)
Extraction by phenol/chloroform and ethanol precipitation
Hammondia heydorniFaecesSucrose purified oocysts10[92]
Grinding with 0.5 mm glass beads in PBS with a dip in liquid nitrogen
Lysis with 2% sarcosyl and 0.5 mg ml−1 pronase E
Extraction by phenol/chloroform and ethanol precipitation
Neospora caninumFaecesNaClO and Sucrose purified oocysts40a[93]
Toxoplasma gondiiLysis in 50 mM Tris, pH 8.0, 100 mM NaCl, 50 mM EDTA, 2% SDS)
Grinding with 2 mm glass beads and further lysis with lysis buffer and proteinase K (1% w/v)
Extraction by phenol/chloroform and ethanol precipitation
Treatment with RNases and re-extraction as described above.
Neospora caninumFaecesSucrose purified oocysts10[94]
Grinding with 0.5 mm glass beads in PBS with a dip in liquid nitrogen
Extraction with DNeasy Tissue Kit (Qiagen)
  • aSensitivity only determined for Neospora caninum.

In environmental samples, soluble substances such as humic acid, clays and polysaccharides are the most important PCR inhibitors [95, 96]. Purification by IMS or FACS is often sufficient to remove inhibitors [88]. Further purification can be achieved by usual DNA extraction procedures (see Table 3) and by incorporating chelating agent-based procedures during extraction to reach low detection limits [69, 89, 97, 98]. From filtration to molecular detection, IMS combined with PCR can recover at least 1 Cryptosporidium oocyst in 5–100 l samples of drinking water and 1–5 oocysts in 20 l of river water samples [97, 98].

5.2.5 Viability and virulence of recovered oocysts

Because usual microscopic analysis and DNA detection provide no information on oocyst viability, tests must be developed to determine the proportion of viable and dead oocysts under various environmental conditions (temperatures, turbidity, salinity). Tests were reported to assess viability of Cryptosporidium sp. and Giardia sp. (oo)cysts, such as inclusion of 4′,6-diamidino-2-phenylindole (DAPI) or exclusion of propidium iodide by viable organisms [99] and mRNA detection by reverse transcription (RT)-PCR [100]. Use of such fluorogenic dyes for T. gondii is not evident, due to probable differences in oocyst wall composition between Cryptosporidium and Toxoplasma, and the presence of an additional sporocyst wall in Toxoplasma. Besides spontaneous autofluorescence of T. gondii oocysts interferes with DAPI fluorescence. Detection of mRNA may be poorly sensitive due to PCR inhibitors in samples. A non-toxic and non-invasive test was recently described for C. cayetanensis oocysts [101]. When they are submitted to a uniform rotating electric field, oocysts exhibit an electrorotation spectrum depending on their viability and sporulation state. This remains to be evaluated for T. gondii in experimental and natural conditions. Efficiency of such techniques should be correlated with bioassays in mice, which are still the reference method to detect viable oocysts. Mice are very sensitive to T. gondii infection and can develop parasites in ascites and/or in tissue cysts in several days or weeks, depending on the dose and virulence of the isolate [11]. Since unsporulated oocysts can be recovered during sample processing, a sporulation time is required before mouse inoculation. Bioassays in mice may separate T. gondii from Hammondia sp. and Neospora sp. according to biological and immunological features of these three closely related coccidia [12, 13]. However, given the fact that a single viable oocyst is able to induce mouse infection, bioassays in mice cannot evaluate the percentage of viable and non-viable oocysts in a suspension. Besides, large amounts of sediments may be problematic for routine detection using mice.

5.3 Detection in soil, fruits and vegetables

Concentration methods cannot be applied for oocyst detection directly from solid matrices due to the large amounts of debris. Recovery is done usually through sample homogenisation, large filtration, standard flotation on sucrose, and/or bioassays in mice [7, 37, 38, 51, 102]. Sediment dispersion with detergent solutions, refinement of flotation procedures with Percoll (Amersham) or cesium chloride (CsCl) density gradients [103], or combining flotation with mAb-based techniques could remove debris, but they may be expensive for routine investigations. According to the sample size, large amounts of PCR inhibitors could influence the molecular detection of oocysts in solid matrices. Laboratory methods have been described to detect C. cayetanensis in fruits and vegetables by molecular tools based on a standard protocol available on the web site of the Food and Drug Administration [104]. They could be adapted to T. gondii detection, with appropriate sequence primers.

5.4 Synthesis of proposed methods

Based on the analysis of current detection methods of Cryptosporidium sp., Giardia sp. and Cyclospora sp. oocysts in environmental samples, we have summarised possible methods for T. gondii detection in water, soil, and food (fruits and vegetables) in Fig. 1. According to oocyst and sample features, use of immunological and/or molecular tools is necessary to recover low numbers of oocysts and to discriminate T. gondii from closely related coccidia. However, bioassays in mice are still required as the reference method to assess oocyst viability since RT-PCR is not yet available. Experimental development of these techniques should take into account the probably low levels of contamination of natural turbid samples.


Possible methods for the detection of T. gondii in environmental samples. a Depending on available sample size. UV, ultraviolet; IMS, immunomagnetic separation; FACS, fluorescence-activated cell sorting; mAbs, monoclonal antibodies; IFA, immunofluorescence assay; PCR, polymerase chain reaction.

6 Concluding remarks

While toxoplasmosis is a veterinary and public health problem, health authorities have no data on T. gondii occurrence in environmental samples because no standard detection methods are available. Assessment of oocyst occurrence in the environment will help answer many questions such as: (i) In which environmental matrix are oocysts more prevalent? (ii) What is the proportion of viable oocysts in the environment under various conditions? (iii) Are there any genetic differences between isolates from intermediate hosts and from the environment? (iv) Are isolates from the environment more virulent than those from intermediate hosts? (v) Which genotype(s) is (are) associated with oocyst-induced toxoplasmosis in humans and in animals? Combining high-resolution typing analysis [1, 105] with detection methods should clarify the key role of oocysts in the epidemiology of toxoplasmosis. Furthermore, it would be possible to minimise contamination, for example, by adapting decontamination procedures for drinking, recreational and irrigation waters according to oocyst resistance and physical features.


The authors wish to thank Bernard Bouteille and Homayoun Riahi for helpful discussion, and Jeanne Cook-Moreau for improving the English of the manuscript. This work was supported by Programme Régional d'Aide à la Recherche Universitaire and Fonds de Croissance Recherche 2002 (Conseil Régional du Limousin).


  1. [1].
  2. [2].
  3. [3].
  4. [4].
  5. [5].
  6. [6].
  7. [7].
  8. [8].
  9. [9].
  10. [10].
  11. [11].
  12. [12].
  13. [13].
  14. [14].
  15. [15].
  16. [16].
  17. [17].
  18. [18].
  19. [19].
  20. [20].
  21. [21].
  22. [22].
  23. [23].
  24. [24].
  25. [25].
  26. [26].
  27. [27].
  28. [28].
  29. [29].
  30. [30].
  31. [31].
  32. [32].
  33. [33].
  34. [34].
  35. [35].
  36. [36].
  37. [37].
  38. [38].
  39. [39].
  40. [40].
  41. [41].
  42. [42].
  43. [43].
  44. [44].
  45. [45].
  46. [46].
  47. [47].
  48. [48].
  49. [49].
  50. [50].
  51. [51].
  52. [52].
  53. [53].
  54. [54].
  55. [55].
  56. [56].
  57. [57].
  58. [58].
  59. [59].
  60. [60].
  61. [61].
  62. [62].
  63. [63].
  64. [64].
  65. [65].
  66. [66].
  67. [67].
  68. [68].
  69. [69].
  70. [70].
  71. [71].
  72. [72].
  73. [73].
  74. [74].
  75. [75].
  76. [76].
  77. [77].
  78. [78].
  79. [79].
  80. [80].
  81. [81].
  82. [82].
  83. [83].
  84. [84].
  85. [85].
  86. [86].
  87. [87].
  88. [88].
  89. [89].
  90. [90].
  91. [91].
  92. [92].
  93. [93].
  94. [94].
  95. [95].
  96. [96].
  97. [97].
  98. [98].
  99. [99].
  100. [100].
  101. [101].
  102. [102].
  103. [103].
  104. [104].
  105. [105].
View Abstract