Toxoplasmosis in Naturally Infected Rodents in Belgrade,Serbia

Introduction

T oxoplasma gondii is one of the most successful parasites on Earth, able to infect an incomparably wide array of hosts, including all warm-blooded animals. The organism's complex life cycle includes an asexual cycle in which encysted parasites circulate between prey and predators. So, infected rodents can serve as a source of tissue cysts for pigs and cats (Dubey and Frenkel 1998, Kijlstra et al. 2008). In addition, rat meat is also welcome on the menu in some countries. The role of domestic and peridomestic rodents in the epidemiology of toxoplasmosis is therefore an issue of investigations around the world (Dubey et al. 1995, Marshall et al. 2004, Kutičić et al. 2005, Hughes et al. 2006, Kijlstra et al. 2008, Murphy et al. 2008), but has never been examined in Serbia. We thus performed a study to assess the prevalence of Toxoplasma infection in wild rodents in an urban setting.

Within the scope of an ongoing large ecological investigation of urban rats as indicators of environmental contamination, 144 brown rats (Rattus norvegicus Berk.) and, incidentally, 12 mice (Mus musculus Linn.) were captured using live animal traps in three localities of Belgrade characterized by poor housing and degraded environment. For this study, brains were obtained from all animals and serum samples from 80 rats (none from mice).

Serology for Toxoplasma-specific IgG antibodies was performed by the modified agglutination test (MAT), as previously described (Desmonts and Remington 1980). Sera were serially twofold diluted starting at 1:25; all sera reactive at ≥1:25 were considered positive (Dubey et al. 1995, Klun et al. 2006).

Aliquots of homogenized brain tissue from all rodents were checked for Toxoplasma cysts by direct microscopic examination and for Toxoplasma DNA by real-time polymerase chain reaction (PCR). All rat brains positive for cysts on direct examination were bioassayed in mice as previously described (Djurković-Djaković et al. 2005); briefly, 850 μL of brain tissue (roughly one-fourth of the brain) together with 100 μL of gentamicin solution was inoculated intraperitoneal in two Swiss-Webster albino female mice each. Six weeks later, mice were sacrificed, blood was collected for serology, and brains were microscopically examined for Toxoplasma cysts.

Complete DNA from rodent brains was extracted with QIAmp DNA mini kit (Qiagen GmbH) according to the manufacturer's instructions. Extracted DNA was resuspended in 100 μL of nuclease-free water and stored at −20°C. The Toxoplasma AF487550.1 gene (529-bp repetitive element), occurring up to 200–300 times in the Toxoplasma genome, was detected using HO1f 5′-AGA GAC ACC GGA ATG CGA TCT-3′ and HO2r 5′-CCC TCT TCT CCA CTC TTC AAT TCT-3′ primers with Taqman probe (10 pmol/μL; 6FAM-ACG CTT TCC TCG TGG TGA TGG CG-TAMRA) (Homan et al. 2000, Reischl et al. 2003, Villena et al. 2004). Amplification was performed on an Eppendorf real-time PCR device (Eppendorf AG). Quality control for molecular diagnostics (Toxoplasma DNA EQA Program) has shown a sensitivity of 1 T. gondii/mL for the real-time PCR protocol performed in our laboratory. To detect PCR inhibitors, DNA from a mimetic plasmid insert (1 pg/mL) was added in a second run to all samples. In case of the presence of inhibitors, bovine serum albumin (Boehringer) was added to a new reaction mixture before the next amplification (Villena et al. 2004). Each amplification run included positive and negative controls. Samples were run in quadruplicate, and only animals that were positive in at least two assays were considered positive. PCR products were analyzed and confirmed on 3% agarose gel stained with ethidium bromide.

The results showed specific IgG antibodies in 22 (27.5%) of the 80 rat blood samples available, of which 13 reacted at a serum dilution of 1:25, 7 at 1:50, and 1 each at 1:100 and 1:200, respectively. Of these, brain cysts were microscopically observed in only two animals (Fig. 1A); curiously enough, both of these reacted in the MAT at only a dilution of 1:25. Brain cysts were also observed in 9 seronegative animals (totally 11, 7.6%; Table 1). Isolation of viable Toxoplasma from both rats and mice with no detectable antibodies to Toxoplasma as tested by MAT has been previously reported (Dubey et al. 1995, Dubey and Frenkel 1998).

OPEN IN VIEWER

FIG. 1.

(A) Toxoplasma gondii cyst in homogenated rat brain tissue (representative example). (B) Products of Toxoplasma gondii real-time polymerase chain reaction on 3% (w/v) agarose gel. Lanes 1 and 2, positive controls (T. gondii RH strain); lanes 3 and 4, negative controls (water and uninfected mouse brain); lanes 5–12, rat brain samples. The amplified fragment is 81 bp (representative example).

Toxoplasma DNA was demonstrated in 15 (10.4%) animals (Table 1 and Fig. 1B), of which 5 (33.3%) had morphologically recognized cysts. However, none of the bioassays performed with cyst-positive rat brains was positive either for cysts or in the MAT, suggesting that the microscopically detected cysts were not viable.

In mice, cysts were detected in 3 (25%), but real-time PCR was positive in even 10 (83.3%) animals, including all 3 with morphologically recognized cysts (Table 1).

Reports on the prevalence of Toxoplasma in naturally infected rodents show a great diversity, particularly in rats. However, this largely depends on the methodology used. Serological studies show a range of virtually 0%–100% infected rats, whereas the seroprevalence of 27.5% found in this study is on the high side of those reported in Europe (Dubey and Frenkel 1998). This may be attributed to the poor hygienic and sanitary conditions of the localities at which the rats were collected, where stray cats and dogs are also frequent.

On the other hand, studies based on the detection of Toxoplasma DNA tend to show relatively higher positivity rates. In a study of 43 rats from urban areas of Manchester, England, even 42.2% were positive (Hughes et al. 2006). We found Toxoplasma DNA in the brains of 10.4% rats, quite similar to the 10.3% rate reported on three organic pig farms in Holland (Kijlstra et al. 2008). The percentage of mice positive for Toxoplasma DNA was remarkably high (83.3%) and included all mice in which brain cysts were observed. However, as the sample was small (12 animals), the results for mice may not be representative. Nevertheless, high rates of Toxoplasma DNA, of 53% and 59%, have been reported in urban house mice (Marshall et al. 2004, Hughes et al. 2006, Murphy et al. 2008).

The reasons why real-time PCR and microscopic detection in rats were not fully concordant may be in part of technical nature. On one hand, microscopic detection has a limited sensitivity, with a lower limit of 10 cysts per 1 mL, which in mice is equivalent to the entire brain, but in rats it is roughly 40 per brain. Obviously, PCR is able to detect lower parasite burdens. On the other hand, PCR from tissue in which cysts were microscopically observed may be negative because of the small amount of rat brain tissue taken for DNA extraction (200 μL), representing ∼7% of the whole brain. This is indirectly supported by the findings in mice, in which all morphologically cyst-positive brains were confirmed by PCR, as a threefold larger proportion of the brain was used for DNA extraction.

Dubey (2009) has reported that identification of Toxoplasma in brain smears is not as sensitive as bioassay. In this respect, it is notable that no bioassay of tissue from rats in our study in which cysts were observed turned out positive. This implies that the microscopically observed cysts were not viable. Isolation of viable Toxoplasma from rats by bioassay is generally infrequent. Iseki et al. (1972) and Frenkel et al. (1995) have each reported one isolation of Toxoplasma out of 11 and 23 seropositive rats examined, respectively. In Illinois farms, Dubey et al. (1995) isolated Toxoplasma from a single rat of the 107 bioassayed (0.9%), of which 6.3% were seropositive. In the Serbia's neighboring country, Croatia, isolation of Toxoplasma by bioassay of brain tissue has been reported in 2 (1.4%) of 142 farm rats and in none of 86 mice, however, of unknown serological status (Kutičić et al. 2005). On the other hand, parasites from rats with proven infection do not always encyst in the mouse brain (Dubey and Frenkel 1998). It seems that the rat immune system induces mutations in Toxoplasma genes that control the parasite's ability to form cysts as well as its pathogenicity (Dubey and Frenkel 1998).

Notwithstanding the above considerations, our results show a considerable level of Toxoplasma infection in rats and mice collected in Belgrade city. The existence of an urban rodent reservoir of infection provides an important link in the epidemiology of toxoplasmosis and suggests that Toxoplasma-infected rats and mice represent a public health risk in this area. As this study was limited to Toxoplasma infection rates in rodents in low-level housing areas, it would be of interest in the future to compare them with those in upper-level quarters. In addition, further studies to determine the extent of rodent predation by cats and prevalence of infection in cats would help to assess the potential risk for humans.