Identification and genetic characterization of Toxoplasma gondii in free-ranging bristle-spined porcupine (Chaetomys subspinosus), a threatened arboreal mammal from the Brazilian Atlantic Forest

Abstract

Background

Strains of Toxoplasma gondii in Brazil have high genetic diversity compared to North America and Europe. The bristle-spined porcupine, Chaetomys subspinosus, is often subject to hunting for human food, but it is not known whether it can be a reservoir of this parasite. The aim of this study was to verify the occurrence of T. gondii in C. subspinosus from southern Bahia, Brazil, and genetically characterize and compare the strains found with those isolated in previous studies of the same region to quantify their genetic diversity by multilocus PCR-RFLP and PCR sequencing.

Findings

Twelve free-ranging C. subspinosus captured in forest fragments of the Una Biological Reserve and adjacent areas were evaluated. Three isolates of T. gondii (TgCsBr01-03) were detected. Two different genotypes were identified by applying multilocus PCR-RFLP with six molecular markers (SAG1, SAG2, SAG3, c22-8, PK1, and Apico). The isolates TgCsBr02 and TgCsBr03 were indistinguishable by this technique. However, the three isolates differed from all the reference strains and from the samples from the same region. Nevertheless, when the six genetic markers were used in multilocus PCR sequencing, all three isolates of T. gondii were different. The phylogenetic analysis revealed a greater genetic distance for TgCsBr01, which was closer to isolates from pigs from the same region, while TgCsBr02-03 was classified in the same lineage and was closer to isolates from sheep from this region.

Conclusions

All the isolates differed from the clonal genotypes of types I, II, and III using both genotyping techniques.

Background

Toxoplasma gondii is an obligate intracellular parasite that infects humans and a variety of warm-blooded animals as the intermediate hosts, with felines as the definitive hosts. T. gondii infections are widely prevalent in humans and animals throughout the world [1].

Molecular studies using PCR-RFLP and microsatellite analysis on isolates of T. gondii in North Africa, Europe, and North America classified the genetic lineages into three types, designated types I, II, III. However, the use of new molecular markers and the study of isolates from South America, especially Brazil, have shown that T. gondii has higher genetic variability [2–4].

Usually, wild animals are reservoirs of T. gondii, and the consumption of their raw or undercooked meat by humans may transport this protozoan [5, 6]. In the Neotropics, consumption of bushmeat from hunting activity by local people is common [7]. In addition to increasing the risk of human contamination by zoonoses, hunting in this biome subjects several species of mammals to direct risk of extinction [8]. This is the case of the bristle-spined porcupine, Chaetomys subspinosus, a rodent species (of the family Erethizontidae) that is an arboreal folivore [9] of medium size [7] and endemic to the Atlantic Forest, whose populations are subject to strong hunting pressure throughout its distribution area, mainly for human consumption [10].

The aim of this study was to verify the occurrence of T. gondii in C. subspinosus from southern Bahia, northeastern Brazil, as well as to genetically characterize and compare the strains found with those isolated in previous studies of the same region to verify their genetic diversity through multilocus PCR-RFLP and PCR sequencing techniques.

Findings

Methods

Collection of samples of biological material

Ethical approval

Experimental samples (brain and blood) [11] were collected from 12 free-ranging adult bristle-spined porcupines from the Una Biological Reserve and adjacent areas, located in southern Bahia, Brazil, from January to November 2013. Of these 12 animals, 8 were captured, sedated [9], the other 4 were found dead by researchers or local residents. All the procedures were performed under the legal approval and consent of the Brazilian Federal Authority (ICMBio, license number: 25184–1; 23468–2 and 27021–1). The proposed study was approved by the ethics committee (CEUA-UESC 024/13).

Molecular diagnosis and genetic characterization of Toxoplasma gondii

DNA extraction

DNA from blood (8 animals) and brain (4 animals) samples was extracted using the commercial kit Easy-DNA™ (Invitrogen). Tachyzoites of the RH strain were diluted at (10⁷/mL) and homogenized, for use as a positive control. DNA was extracted and performed according to Bezerra et al. [3]. Samples stored at −20 °C.

Diagnosis by PCR

T. gondii was detected by polymerase chain reaction (PCR) which amplified a fragment of 529 bp utilizing the primers Tox4 Forward (CGCTGCAGGGAGGAAGACGAAAGTTG) and Tox5 reverse (CGCTGCAGACACAGTGCATCTGGATT) [12].

PCR-RFLP for genetic characterization

The genotypes of T. gondii isolated were determined by means of multilocus PCR-RFLP with six genetic markers: SAG1, SAG2, SAG3, c22-8, PK1, and Apico. The amplification reactions were performed according to Bezerra et al. [3]. The digestions were carried out according to Su et al. [13]. The patterns of the DNA bands of the samples were compared with the genotypes deposited in ToxoDB (http://toxodb.org/toxo/).

DNA sequencing

The products from nested PCR were purified using PureLink™ (Invitrogen) and sequenced for six genetic markers (SAG1, SAG2, SAG3, c22-8, PK1, and Apico) using the automatic sequencer ABI-PRISM 3100 Genetic Analyzer (Applied Biosystems). As positive controls, nested PCR products of RH (type I), PTG (type II), and CTG strains (type III) were sequenced. Nucleotide sequences determined in this study were assembled in contigs using CAP3. The sequences of T. gondii were aligned with ClustalW (version 1.83; [14]), manually corrected using BioEdit Sequence Alignment Editor, and compared with 10 reference sequences of T. gondii available at NCBI (http://www.ncbi.nlm.nih.gov/bioproject/). These strains were GT1 (PRJNA16727), ME49 (PRJNA28893), VEG (PRJNA19097), FOU (PRJNA61561), MAS (PRJNA61545), VAND (PRJNA60839), RUB (PRJNA61119), p89 (PRJNA61547), TgCATBr5 (PRJNA61551), and TgCATBr9 (PRJNA61549). For the Apico marker, the sequences were aligned with the T. gondii apicoplast complete genome (U87145.2). All the sequences were compared with sequences available at ToxoDB.

Phylogenetic and statistical analysis

The phylogenetic analysis was performed using MEGA version 6 by means of the neighbor-joining algorithm, and the distances were computed using the Tajima-Nei method. The stability of branches was assessed after bootstrapping with 500 replicates. To verify the distance between samples of the same geographical origin, eight T. gondii isolated from pigs from southern Bahia (TgPgBr06, TgPgBr08, TgPgBr09, TgPgBr11, TgPgBr12, TgPgBr13, TgPgBr15, and TgPgBr16) [3] and three T. gondii isolated from sheep from southern Bahia (#54, #124, and #127) [4] were included. Tajima’s test of neutrality [15] was used to compare the number of segregating sites with the nucleotide diversity of the DNA sequences.

Results

Genetic characterization by multilocus PCR-RFLP

Three (25 %) animals were positive for T. gondii based on the biological samples. The positive samples were from three different and adjacent forest fragments within the Una Biological Reserve: TgCsBr01, TgCsBr02 and TgCsBr03. The greatest distance between the positive records was approximately 6 km. The application of PCR-RFLP with six genetic markers (SAG1, SAG2, SAG3, c22-8, PK1, and Apico) revealed two genotype groups in the three isolates (Table 1).

Table 1 Multi-locus genotypes of Brazilian Toxoplasma gondii isolates by PCR-RFLP

Genetic characterization by multilocus PCR sequencing

A total of 1,604 positions were used in the alignment to calculate the evolutionary rate among the concatenated sequences of the samples TgCsBr01, TgCsBr02, and TgCsBr03, which was indicated through Tajima’s relative rate test. There were 1,461 identical sites and no divergent sites between the three samples. The null hypothesis of equal rates among the lineages was rejected (P < 0.05), and one of the three samples (TgCsBr01) was considered to be from a different lineage.

The samples differed from the 10 strains of T. gondii and the apicoplast genome, presenting a mean of 120.6 DNA polymorphisms (5.2 %), including insertion, deletion, transition, and transversion, distributed over the different genetic markers (Table 2).

Table 2 Genotyping by PCR-RFLP and number of polymorphisms at six genetic loci detected by PCR sequencing

Through Tajima’s D test, which measures the distribution of allele frequency based on the data of nucleotide sequences, 1,870 bases could be aligned, showing 388 segregating sites. The analysis showed a positive result (0.372), which was indicative of a low number of low-frequency polymorphism among all the strains of T. gondii used in this analysis.

Phylogenetic analysis

The phylogenetic analysis of the DNA sequences by the neighbor-joining method demonstrated that TgCsBr02 and TgCsBr03 were grouped in a group distinct from TgCsBr01. TgCsBr02 and TgCsBr03 were considered a sibling group, defined by an apomorphy of 99 % of the bootstrap replicates. TgCsBr01 presented the largest branch length, which was proportional to the amount of polymorphisms. The isolates TgCsBr02 and TgCsBr03 were the closest to the reference strains GT1 (type I), ME49 (type II), and VEG strain (type III), but were grouped in a distinct group with 15 % of the bootstrap replicates (Fig. 1).

Fig. 1

Phylogram of Toxoplasma gondii strains as determined by sequence analysis of the five genomic markers

Discussion

This was the first study that identified and genetically characterized T. gondii from C. subspinosus. None of the samples were classified as a clonal genotype of type I, II, or III (Table 1) or as a main Brazilian clonal genotype (BRI, BRII, BrIII, and BRIV) defined by Dubey et al. [16] and Pena et al. [2]. The samples were also genetically distinct from genotypes of T. gondii previously isolated from pigs and sheep of the same geographic region of Brazil (southern Bahia), as described by Bezerra et al. [3] and Maciel et al. [4] (Table 1). Moreover, the samples did not combine with any genotype deposited in ToxoDB.

However, the phylogram determined by the analysis of the chromosomal genetic marker sequence revealed that the isolates TgCsBr02 and TgCsBr03 were closer to the ovine strains of the same area of study (#54 and #127). TgCsBr01 shared a monophyletic group in 99 % of the replicates with swine isolates from the same geographical region (TgPgBr06, TgPgBr07, TgPgBr08, TgPgBr09, TgPgBr10, TgPgBr11, TgPgBr12, TgPgBr13, TgPgBr14, TgPgBr15, and TgPgBr16), which suggests that it may belong to a close lineage that inhabits this region. However, the genetic characterization of the parasite carried out by means of multilocus PCR-RFLP and DNA sequencing techniques indicated a high genetic diversity of the parasite in the region (Tables 1 and 2; Fig. 1).

The genetic characterization by PCR-RFLP revealed two genotypes in the three strains. Nevertheless, it was verified that the isolates of C. subspinosus possessed many of the same alleles present in other Brazilian genotypes, although the alleles had distinct segregation among the loci examined, thereby characterizing it as a new genotype. The multilocus PCR sequencing indicated that the three isolates from the C. subspinosus samples were distinct, and the genetic variability between the samples was enough to classify them into different genotypes in the phylogenetic analysis. This was indicated by Tajima’s relative rate test, which calculated the equality of evolutionary rates between the concatenated sequences generated by multilocus PCR sequencing (Fig. 1).

The most polymorphic genetic markers were c22-8 and Apico, with alignments presenting 90 and 95 % identity with the TgCatBr5 and apicoplast genomes, respectively (Table 2). Despite this high degree of polymorphism, the samples were grouped in the clonal genotype type III by the PCR-RFLP technique (Table 1). These regions may be considered effective to distinguish isolates of clonal types I, II, and III, though this complicates the grouping of Brazilian isolates.

Few studies have genetically characterized T. gondii in wild mammals, and most research in Brazil is related to serological techniques in these animals [5, 6, 17, 18]. Most isolates of T. gondii genotyped in Brazil are from domestic animals, including chickens, cats, dogs, sheep, goats and pigs. The studies that have genetically characterized isolates of T. gondii from wild animals have also reported a high genetic diversity and have contributed to the elucidation of the Brazilian genotype network [19–21]. Silva et al. [6], upon detecting T. gondii in armadillos in Brazil, claimed that these wild animals can be a source of transmission of the parasite to humans, mainly due to the rural people’s habit of eating the meat of these animals.

Conclusions

Little is known about the genetic variability of isolates of T. gondii in wild mammals in Brazil. Knowing the genetic similarities or differences of T. gondii between different animal populations is necessary to understand the transmission of the parasite.