INTRODUCTION
Toxoplasma gondii infections are widely prevalent in human beings and other animals worldwide (Dubey and Beattie, Reference Dubey and Beattie1988). Humans become infected post-natally by ingesting tissue cysts from undercooked meat, consuming food or drink contaminated with oocysts, or by accidentally ingesting oocysts from the environment. Until recently, T. gondii was considered to have low genetic diversity. We have recently found that the isolates of T. gondii from Brazil are biologically and genetically different from those in North America and Europe (Dubey et al. Reference Dubey, Graham, Blackston, Lehmann, Gennari, Ragozo, Nishi, Shen, Kwok, Hill and Thulliez2002; Lehmann et al. Reference Lehmann, Marcet, Graham, Dahl and Dubey2006; Dubey et al. Reference Dubey, Applewhaite, Sundar, Velmurugan, Bandini, Kwok, Hill and Su2007a). Toxoplasma gondii isolates from asymptomatic chickens from Brazil were more pathogenic to mice than isolates from Europe or North America, irrespective of the genotype (Dubey et al. Reference Dubey, Patitucci, Su, Sundar, Kwok and Shen2006a). Therefore, strains isolated from chickens in Brazil possess greater genetic diversity than those identified in Europe and North America.
Cats are essential in the life cycle of T. gondii because they are the only hosts that can excrete the environmentally-resistant oocysts in nature (Dubey and Beattie, Reference Dubey and Beattie1988). We have started to study genetic diversity of T. gondii isolates from cats but it is difficult to obtain samples from feral domestic cats (Felis domesticus). Until now, data were obtained from domestic cats from Brazil (Dubey et al. Reference Dubey, Navarro, Sreekumar, Dahl, Freire, Kawabata, Vianna, Kwok, Shen, Thulliez and Lehmann2004; Pena et al. Reference Pena, Soares, Amaku, Dubey and Gennari2006), The Peoples Republic of China (Dubey et al. Reference Dubey, Zhu, Sundar, Zhang, Kwok and Su2007c), and Colombia (Dubey et al. Reference Dubey, Su, Cortés, Sundar, Gomez-Marin, Polo, Zambrano, Mora, Lora, Jimenez, Kwok, Shen, Zhang, Nieto and Thulliez2006b). In the present study, the seroprevalence and the isolation and genetic characterization of T. gondii from feral cats on St Kitts, West Indies are reported.
MATERIALS AND METHODS
Naturally-infected cats
The St Kitts island is located in the Eastern Caribbean at 17 20 North, 62 45 West with habitats ranging from dry areas to rainforests. The island is 179 km2 with a population of approximately 35 000. In total, 96 feral cats (46 males, 50 females; 14 <6 months of age, and 82 adults) within 10 km radius of Basseterre, St Kitts were captured from their original colonies in cages and identified by ear notching during 2005 and 2006. These cat colonies represent the most populated region of the island in and near the capital, Basseterre. During the day 2/3 of the island population commute to work in Basseterre and after several hours return to their homes in the rural areas on the island. All procedures involving handling, sample collection, surgery and euthanasia were performed according to the Institutional Animal Care and Use Committee (IACUC) from Ross University, School of Veterinary Medicine. Cats were housed and fed ad libidum at the Ponds Veterinary Hospital, a local veterinary clinic owned by Dr Burnell Nisbet. For easier handling and safety precautions all feral cats were anesthetized using 50% ketamine, 50% zylazine solution (1 ml/cat) administered intramuscularly. Approximately 3 ml of blood were obtained from the jugular vein, labelled, centrifuged, and serum stored at −20°C. Feces from the rectum of 51 cats were obtained for parasitological examination. All cats were tested for Feline Immunodeficiency Virus (FIV) and Feline Leukemia Virus (FeLV) infections using the SNAP Combo Test kits (Idexx Laboratories, Westbrook, Maine, USA). None of the 96 cats were positive for FeLV. Fifteen of the 96 cats were positive for FIV and these 15 cats were euthanized using an overdose (>50 mg/kg) of pentobarbital. All cats were clinically examined and their general health status determined. Most of the FIV-positive animals were adult males showing scratches and bites, indicating a more aggressive behaviour, but in a good health condition otherwise. The remaining 81 cats were de-wormed, vaccinated, spayed/neutered and returned to their original colonies approximately 5 days after surgery.
Samples of sera from 96 cats, feces from 51 cats, and tissues from 10 cats were shipped with cold packs from St Kitts to Beltsville, Maryland for T. gondii evaluation.
Serological examination for T. gondii
Two-fold serial dilutions were made (1:10 to 1:1280) and tested with a modified agglutination test (MAT), as described previously (Dubey and Desmonts, Reference Dubey and Desmonts1987).
Bioassay of feline tissues for T. gondii infection
Brain, heart, and tongue of each of the 10 cats were bio-assayed for T. gondii infection in out-bred female Swiss Webster mice obtained from Taconic Farms, Germantown, New York, as described by Dubey et al. (Reference Dubey, Graham, Blackston, Lehmann, Gennari, Ragozo, Nishi, Shen, Kwok, Hill and Thulliez2002). Each tissue (20–50 g) was homogenized individually, digested in an acidic pepsin solution, neutralized, centrifuged (Dubey, Reference Dubey1998); the homogenate was inoculated subcutaneously (s.c.) into 5 mice (Table 1). The recipient mice were examined for T. gondii infection. Tissue imprints of mice that died were examined for T. gondii tachyzoites or tissue cysts. Survivors were bled on day 40–42 post-inoculation (p.i.) and a 1:25 dilution of serum from each mouse was tested for T. gondii antibodies with the MAT. Mice were killed 6 weeks p.i. and brains of all mice were examined for tissue cysts as described (Dubey and Beattie, Reference Dubey and Beattie1988). The inoculated mice were considered infected with T. gondii when tachyzoites or tissue cysts were found in tissues.
Isolation of Toxoplasma gondii from feral cats from St Kitts
a No. of mice infected with T. gondii from 4 mice inoculated with feline tissues.
b Mice were killed accidentally 34 day p.i. and not examined for T. gondii infection.
c One mouse died on 16 day p.i.
Bioassay of feline feces for T. gondii oocysts
Feces (1–5 g) from the rectum of 51 cats were floated in sugar solution (sp. gr. 1·18), and the float was examined microscopically for oocysts. Approximately 0·5 ml of the float from the very top of the tube was mixed with 10 ml of 2% aqueous sulfuric acid, aerated at room temperature for 1 week and sent to Beltsville for bioassay. The acid in the sample was neutralized, centrifuged, and inoculated orally into mice (Dubey et al. Reference Dubey, Hill, Jones, Hightower, Kirkland, Roberts, Marcet, Lehmann, Vianna, Miska, Sreekumar, Kwok, Shen and Gamble2005).The recipient mice were examined for T. gondii infection.
Genetic characterization
Toxoplasma gondii DNA was extracted from the tissues of all 46 infected mice (Table 1) and strain typing was performed using 10 PCR-RFLP genetic markers SAG1, SAG2, SAG3, BTUB, GRA6, c22-8, c29-2, L358, PK1 and Apico (Dubey et al. Reference Dubey, Patitucci, Su, Sundar, Kwok and Shen2006a, Reference Dubey, Zhu, Sundar, Zhang, Kwok and Su2007c; Su et al. Reference Su, Zhang and Dubey2006). To confirm mixed infections, PCR products were first sequenced from both ends using sequencing primers (forward and reverse). If the DNA sequence chromatogram showed 2 nucleotide peaks at the expected single nucleotide polymorphism (SNP) site, then the PCR products were cloned by TA cloning using pGEM – T Easy vector system (Promega, Madison, Wisconsin) following the manufacturer's instructions. Ten to 25 clones for each isolate are picked to expand in 1 ml of LB broth with 100 μg/ml ampicillin. These clones were genotyped again and representative clones were sequenced to confirm the existence of 2 different DNA sequences. Three loci including SAG2, c22-8 and L358 were sequenced for the representative isolates from each cat that had potential mixed infection based on PCR-RFLP results. Here, the SAG2 is the marker based on the 5′-end of the gene sequence (Su et al. Reference Su, Zhang and Dubey2006). Reference strains RH88, PTG and CTG for Type I, II and III lineages, respectively, were also included for sequencing. The primers used for sequencing are: SAG2-SqF, TAGCTTTCAAGACCGCACCT and SAG2-SqR, CTGCTTGCGATTCTGTGTGT for SAG2; c22-8SqF, AAGGATCGGGGAAAGTGTCT and c22-8SqR, GCGAACCTTCTGTCATCTCC for c22-8; and L358-SqF, ATGTCCTCTTTCTGCCTTCG and L358-SqR, GGAGAAAGCGAAACCTTCCT for L358.
RESULTS
Antibodies to T. gondii were found in the sera of 71 of 96 (73·9%) of cats with titres of 1:10 in six, 1:20 in six, 1:40 in seven, 1:80 in three, 1:160 in 10, 1:320 in 13, 1:640 in nine, and 1:1, 280 or higher in 17.
Toxoplasma gondii was isolated from tissues of 7 of the 10 cats; from the hearts of 6, tongues of 5, and the brains of 3 cats (Table 1). All 7 isolates of T. gondii were avirulent for mice (Table 1). Toxoplasma gondii was not isolated from the feces of cats from St Kitts.
Genotyping of the 7 isolates revealed 4 genotypes, including clonal Type II, III and 2 unique genotypes. Tissues from 5 of the 7 cats, when inoculated into mice, gave rise to infections with more than 1 strain of T. gondii (Table 2). Isolates TgCatStK1 and TgCatStK7 each had a unique genotype that is not found from South America isolates studied to date. TgCatStK2, TgCatStK4 and TgCatStK6 each had a Type III genotype with the possibility of mixed infection with Type II genotype. TgCatStK3 clearly showed a mixed infection of Type II and III genotypes. TgCatStK5 showed the possibility of a mixed infection with Type II and III genotypes.
Genotyping of Toxoplasma gondii isolates from feral cats from St Kitts, West Indies
a Isolates derived from B=Brain, H=Heart, T=Tongue.
b SAG2 marker based on 5′- and 3′-ends of the gene sequence (Howe et al. Reference Howe, Honore, Derouin and Sibley1997).
c No. of mice with the same genotype.
d A new SAG2 marker based on the 5′-end of the gene sequence (Su et al. Reference Su, Zhang and Dubey2006).
e ND, no data.
f Not enough DNA for genotyping from 1 of the 4 infected mice.
To confirm mixed infection, 1 representative isolate from each cat with a potential mixed infection was sequenced at 3 loci, including SAG2, c22-8 and L358. DNA sequencing data from at least 2 of the 3 loci confirmed the PCR-RFLP data. An example of the DNA chromatogram trace data for TgCatSk2 at locus SAG2 is summarized in Fig. 1. It is shown that 5 of the 7 cat isolates, including TgCatStk2, 3, 4, 5 and 6 have mixed infection of 2 genotypes. The mixed infection was further confirmed by TA cloning of PCR products (marker SAG2 or c22-8), genotyping 10–25 individual clones for each isolate and then sequencing representative clones. Four of the above 5 isolates, including TgCatStk 2, 3, 4 and 5, were confirmed to have 2 different sequences. TgCatStk6 was not done due to an insufficient amount of DNA samples. Figure 2 shows the results of genotyping individual clones from TA cloning of TgCatStk4 at the SAG2 locus.
Summary of DNA trace data that contain the mixed nucleotides at SAG2 locus. The data were based on the forward and reverse sequences of SAG2 locus. The polymorphic sites were indicated by the letters of the nucleotide. Reference strains RH88, PTG and CTG represent the clonal Type I, II and III lineages, respectively. TgCatSk2 with potential mixed infection showed 2 nucleotides at 5 positions that matched exactly to the polymorphic sites among the 3 reference strains.
Genotyping of individual clones from TA cloning of TgCatStk5 at the SAG2 locus. Sample A49 is the direct PCR product showing mixed genotypes (types II and III). Samples 6, 7, 8, 14 and 15 are clones derived from A49 by TA cloning. Sample 6 shows a type III allele, whereas samples 7, 8, 14, and 15 have type II alleles. DNA sequencing of samples 6 and 14 confirmed the mixed infection of types II and III alleles. MK is the DNA marker. RH88, PTG and CTG are reference type I, II and III lineages, respectively.
DISCUSSION
The results indicate that the environment on St Kitts is highly contaminated with T. gondii oocysts already shed by the sero-positive cats. Cats shed T. gondii oocysts only for a short time (<1 week) and at any given time less than 1% were shedding oocysts (Dubey, Reference Dubey2004). Cats usually develop antibodies to T. gondii 1–2 weeks after they have shed oocysts (Dubey, Reference Dubey2004). This may be the reason why oocysts were not detected in feces of any of the 51 cats examined in the present study. Infected cats can release millions of T. gondii oocysts into the environment and people become exposed by ingesting the oocysts or by eating undercooked meat from animals which have ingested oocysts and developed tissue cysts. Although there are no published data for St Kitts, our study suggests that toxoplasmosis might be an important disease on this island and health workers on this and neighbouring islands should be alerted to this possibility.
In the present study T. gondii was isolated from 7 of the 10 cats that were also infected with the FIV. These 7 cats appeared to have no clinical signs and there is no evidence that the 70% isolation rate of T. gondii from sero-positive cats with dual infection with T. gondii and FIV is different than cats in the general population (Dubey et al. Reference Dubey, Navarro, Sreekumar, Dahl, Freire, Kawabata, Vianna, Kwok, Shen, Thulliez and Lehmann2004; Pena et al. Reference Pena, Soares, Amaku, Dubey and Gennari2006). Viable T. gondii was isolated from 68·5% of 54 sero-positive (Dubey et al. Reference Dubey, Navarro, Sreekumar, Dahl, Freire, Kawabata, Vianna, Kwok, Shen, Thulliez and Lehmann2004) and 66·2% of 71 sero-positive (Pena et al. Reference Pena, Soares, Amaku, Dubey and Gennari2006) cats from Brazil. However, cats from Brazil were not tested for the FIV infection. The FIV is morphologically and biologically similar to the human immunodeficiency virus (Levy et al. Reference Levy, Ritchey, Rottman, Davidson, Liang, Jordan, Tompkins and Tompkins1998); there are no indications that the FIV infection modifies the course of latent T. gondii infection in cats (see Dubey and Carpenter, Reference Dubey and Carpenter1993).
Until recently, T. gondii was considered to have an affinity for encystment in neural tissue but this assumption is based on infections in mice. Results of the present study indicate that the muscular tissues in cats are more heavily parasitized with T. gondii than the brain, which supports earlier findings (Dubey et al. Reference Dubey, Navarro, Sreekumar, Dahl, Freire, Kawabata, Vianna, Kwok, Shen, Thulliez and Lehmann2004, Reference Dubey, Su, Cortés, Sundar, Gomez-Marin, Polo, Zambrano, Mora, Lora, Jimenez, Kwok, Shen, Zhang, Nieto and Thulliez2006b).
Genotyping of the 7 cat isolates from St Kitts identified 4 genotypes. Two genotypes are unique and were not identified from isolates collected from a variety of hosts in South America including Brazil, Colombia, Chile, Costa Rica, Nicaragua and Guyana (see Dubey et al. Reference Dubey, Applewhaite, Sundar, Velmurugan, Bandini, Kwok, Hill and Su2007 a, Reference Dubey, Sundar, Gennari, Minervino, Farias, Ruas, dos Santos, Cavalcante, Kwok and Sub). The other 2 genotypes are identical to the clonal Type II and III lineages that are predominant in North America and Europe. This study showed that there is potentially a high percentage of mixed infections among the feral cats. To identify mixed infections with both Type II and Type III genotypes with ultimate certainty, the actual parasites from the infected tissues need to be isolated, cloned and re-examined at all PCR-RFLP loci. Unfortunately, these infected tissues are no longer available at this point and further experiments are not possible.
This project was funded in part by Ross University School of Veterinary Medicine (RUSVM). We thank the RUSVM Diagnostics Research Laboratory for preparing all samples, and Cassandra Benjamin, Jaime Thurk, Tanya Miller, Gisele Roy and Tiffany Caudill for technical help.
