Index References

Canine transmissible venereal tumor (CTVT) is the only neoplasm that can be spread horizontally among dogs and it mainly affects the external genitalia in both genders. In 1876, Novinski was the first to report the transmissible characteristics of CTVT and ever since, this neoplasm has been recognized as a contagious tumor (Karlson and Mann, 1952). Thus, CTVT does not originate by the malignant transformation of a cellular type in affected individuals, but instead, the tumor cells are transmitted between dogs in a way that resembles/similar parasitic transmission. Hence, tumor cells themselves are the infectious agents and therefore are genetically distinct from their host (Frank, 2007; Murgia et al., 2006; Rebbeck et al., 2009). Recently, it was discovered that CTVTs throughout the world, originated from the same cellular clone (from wolves) probably 65,000 years ago (Rebbeck et al., 2009). This finding allowed to establish the monophyletic origin of the disease. In accordance with its clonal origin, all CTVT cells have a genomic rearrangement produced by the insertion of a long interspersed nuclear element (LINE) near the 5’ end of c-Myc (Liao et al., 2003). Furthermore, CTVT karyotype includes 58 to 59 chromosomes, with 16 or 17 rearranged as metacentric, instead of the 78 acrocentric chromosomes characteristically found in dogs (Canis familiaris) (Makino, 1963; Murray et al., 1969; Breen, 2008). Despite CTVT’s monophyletic origin, several reports have found the presence of genetic diversity in this tumor. Currently, CTVT throughout the world can be grouped in two genetic lineages identified by different mitochondrial DNA (mtDNA) haplotypes (Murgia et al., 2006). In this sense, clones belonging to both lineages have been transmitted to both ancient and modern breeds of dogs (vonHoldt and Ostrander, 2006).

The aims of this study were to determine whether Mexican CTVT haplotypes are genetically diverse from CTVT haplotypes present in other countries, as well as exploring the genetic relationship among tumors in Mexico. To achieve this, we used single nucleotide polymorphisms (SNPs) found within a fragment of mtDNA (D-loop region) as a genetic marker in 50 CTVT tumors and their corresponding blood samples. The data of matched tumor tissues and blood samples are shown in Table 1.

View Details

Table 1 Dogs providing fresh tumor and blood samples

All tumors sequenced in this study were initially diagnosed by histopathology. Diagnosis was confirmed by detection of LINE insertion in c-Myc gene using PCR, following the protocol proposed by Liao et al. (2003). In order to genotype mtDNA, we extracted DNA from both tumors and hosts using the Wizard Genomic DNA Purification Kit and the GE Illustra Blood GenomicPrep Mini Spin Kit, respectively (Promega). Next, using the primers H15422/H15710 (Murgia et al., 2006), we amplified a 290-bp fragment of mtDNA control region in 100 samples (50 CTVTs and 50 blood samples of the corresponding host). The PCR was performed in a final volume of 25 μl containing 20 ng of DNA, 10 nM dNTPs, 1 U Taq DNA Flexi polymerase, 5 μl of Green Buffer and 1.5 mM of MgCl₂ (Promega). The amplification conditions were 95°C for 10 min; 30 cycles of 94°C for 40 sec, 57°C for 40 sec, and 72°C for 45 sec, followed by a final extension at 72°C for 5 min. The amplified fragments were visualized on a 3% agarose gel and purified using the QIAquick Gel Extraction Kit (Qiagen). Finally, the purified fragments were sequenced using an ABI Prism 3100 Genetic Analyzer (Applied Biosystems).

In accordance with previous studies (Murgia et al., 2006; Rebbeck et al., 2009), none of the 50 analyzed tumor sequences, matched the sequence of their respective hosts. Out of the 50 CTVTs analysed in the present study, thirty-nine had identical sequences and were included into a single haplotype labelled as TVT1. Within the remaining nine haplotypes, TVT5 and TVT4 included two sequences each, while the rest of the haplotypes (7) included only one sequence each. Therefore, ten new haplotypes were identified in Mexican CTVT samples (Table 2). These can now be added to the 41 haplotypes that have been identified throughout the world (Murgia et al., 2006; Rebbeck et al., 2011).

View Details

Table 2 Variable sites among the 10 haplotypes identified from mtDNA control region sequences of 50 samples of Mexican CTVT

Haplotypes obtained from CTVT in Mexico were compared with CTVT haplotypes found in others countries including Kenya, Brazil, Italy, India, Spain, Turkey, Israel, South Africa, Thailand, Greece, Malaysia, Chile and USA (Murgia et al., 2006; Rebbeck et al., 2011). The geographic distribution and the accession number of the sequences of all tumors are shown in Table 3. We performed all molecular and phylogenetic analyses using MEGA 5 software (Tamura et al., 2011). Through multiple alignments of the sequences, we visualized the relationships among CTVT sequences from different countries and constructed a statistical parsimony network 4.5.1.6 (Bandelt et al., 1999).

View Details

Table 3 CTVT haplotype mtDNA from different countries

When we combined all haplotypes to obtain genetic relations of CTVT (Fig. 1), the maximum genetic distance (0.26) occurred between haplotype TVT3 from Mexico and 12ER from Turkey. The network obtained, including all CTVTs reported in different countries, was organized into two main clusters (I and II), which is consistent with the data reported by Murgia et al. (2006). Cluster I consisted of 43.2% of the sequences, distributed into 22 haplotypes and cluster II consisted of 56.8% of the sequences distributed into 29 haplotypes. The most frequent haplotype found in cluster I was 4AR, which includes 31 sequences (three from Turkey, six from Spain, one from Brazil, nine from Italy, two from Kenya, four from South Africa, three from Greece and three from Malaysia). However, this particular haplotype was not found within our Mexican samples.

View Details

Fig. 1 Haplotypes found in CTVT from different countries including Mexico. The diameter of each circle is proportional to the number of tumor samples; each branch represents one base pair change, with red dots representing intermediates not found in the tumor samples analyzed. The outlined boxes indicate mtDNA clusters.

The most frequent haplotype found in cluster II, was labelled as TVT1, it consisted of 53 sequences from Mexico (39 sequences reported here and 13 reported in Rebbeck et al. (2011)) and one from Chile (this study).

Since TVT1 was the most ubiquitous haplotype within our Mexican samples, we suggest that it could be the origin of the different haplotypes present in Mexico. In addition, the results of our study suggest that most Mexican haplotypes are genetically related to those found in India, USA, Greece, Israel and Thailand. Furthermore, CTVT haplotypes from the American Continent (Mexico, USA, Chile and Brazil) are closely related to Asian haplotypes.

Given that CTVT arose approximately 65,000 years ago (Rebbeck et al., 2009), prior to the migration of humans (and consequently dogs), from Asia into the American Continent (approximately 16,000 years ago) (Leonard et al., 2002), we speculate that dogs migrating into the new world, brought CTVT. However, further studies are needed to test this hypothesis.

In the clade I network, we observed only three Mexican haplotypes that were associated with haplotypes from Europe. However, a larger number of CTVT samples are needed to explain the relationship of Mexican haplotypes with European ones. In this study we found that CTVT samples of both ancestral lineages are circulating in Mexico (Fig. 1).

A further contribution of this study is the identification of 10 new haplotypes that complement previously proposed networks on CTVT genetic relationships (Murgia et al., 2006). Specifically, the network presented by Murgia et al. (2006), contains a non-identified haplotype between the 29GR sample from Brazil and the central node of the cluster I. In the network developed in this study, the missing haplotype sequence was identified as TVT8. Additionally, another formerly unidentified haplotype between haplotypes 8CR and 8F was identified as 56C from South Africa (Fig. 1).

Recently it has been proposed that during the course of CTVT evolution, tumor cells may acquire mitochondria from their hosts, this is supported by the topology of a parsimony tree constructed by Rebbeck et al. (2011). In order to construct the tree, the authors included sequences from both CTVTs and hosts, and as a result, they obtained two distinctive clusters.

In the light of this recent discovery, we constructed a tree including host and CTVT sequences. The resulting parsimony tree is also divided in two main clusters and in according with Rebbeck et al. (2011) we observed mtDNA haplotypes of dogs present in any CTVT samples (Data not show).

Finally, in a previous study, in which CTVT samples from Yucatan State, Mexico were characterized with microsatellites, no genetic diversity was found (Rebbeck et al., 2009). In the present study we reported to mtDNA instead and analysed samples from Mexico State and Veracruz. In both cases we observed genetic diversity. Therefore, we believe that further studies on CTVT genetic diversity would benefit from using mtDNA as a molecular marker. Characterizing additional CTVT haplotypes would allow us to conduct molecular epidemiology studies of this tumor in order to better understand its origin and transmission, both locally and across the world.

The authors thank the Universidad Autónoma del Estado de México (UAEMex) for funding the present study. We would also like to thank Dr. Claudio Murgia for providing access to mtDNA sequences. Finally, Linda G. Bautista would like to thank Consejo Nacional de Ciencia y Tecnología (CONACYT) for awarding her a postgraduate fellowship.