Clinical-dermatological, histological abnormalities and prevalence of Trypanosoma caninum and Leishmania infantum in dogs from Midwest region of Brazil.

Leishmania infantum is a trypanosomatid that causes parasitic dermatopathy in dogs. Trypanosoma caninum is another trypanosomatid, which infects the skin of dogs, although cutaneous abnormalities are absent. This study aimed to investigate the occurrence of T. caninum infection and its associated cutaneous and histological changes and compare it with the occurrence of L. infantum infection in dogs. The study included 150 dogs, of which T. caninum infection was identified in 3 (2%) and L. infantum infection in 15 (10%) of them, with no association (p>0.05) of these infections with the breed, gender, age, or cutaneous abnormalities. The cutaneous abnormalities were based on 1 (4.8%) and 12 (57.1%) dogs infected by T. caninum and L. infantum, respectively. The dermatohistopathological abnormalities in the dogs infected with T. caninum included mild perivascular lymphohistioplasmacytic infiltrates in the clinically asymptomatic ones, while in those with dermatological abnormalities, acanthosis, epidermal orthokeratotic hyperkeratosis, melanomacrophages, and co-infection with Microsporum sp. and Trichophyton sp. were observed. InL. infantum infected, the histopathological findings included chronic granulomatous inflammatory infiltrates and structures compatible with amastigotes. Despite the low frequency of T. caninum infection, our findings suggest that this trypanosomatid, unlike L. infantum, does not cause any macroscopic skin abnormalities.


Introduction
Among parasitic dermatoses, those caused by the protozoa of the Trypanosomatidae family, especially of the Leishmania genus, are relevant to dogs residing in endemic areas of Brazil . Although many Trypanosoma species affecting dogs have been described (Hoare, 1972), there is no description of the cutaneous changes caused by these protozoa. However, in 2009, a new Trypanosoma species, named Trypanosoma caninum, parasitizing the skin of a dog co-infected with Leishmania braziliensis was described (Madeira et al., 2009), suggesting that dermatological changes may be associated with this parasite.
Little is known about the development cycle, tropism, and pathogenicity of T. caninum (Madeira et al., 2009). T. caninum has been isolated only from the skin so far, a factor that is unusual of the Trypanosoma genus, and it shares the same reservoir with L. infantum, the infectious agent of visceral leishmaniasis (VL). It has been isolated from several regions of Brazil (Alves et al., 2012;Oliveira et al., 2015). These aspects raise some public health concerns, especially the risk of erroneous diagnoses of LVC in areas of sympatry of L. infantum and T. caninum (Almeida et al., 2011;Alves et al., 2012).
The aspects described above, and the limited information available related to the clinical aspects and the possible dermatological alterations associated with T. caninum infection in dogs motivated this study. We aimed to investigate the frequency of T. caninum infection in dogs selected at the University Veterinary Hospital of Cuiabá, Mato Grosso, Brazil, and examine possible dermatological and histological abnormalities in T. caninuminfected dogs and compare it to those seen in dogs infected with L. infantum.

Dogs
This study included 150 dogs, selected while attending clinic and dermatological at the Veterinary Hospital (HOVET) of the Federal University of Mato Grosso (UFMT) campus Cuiabá, Mato Grosso, from May 2015 to May 2016. As inclusion criteria, dogs with a history of dermatological alterations (symptomatic) and healthy dogs (asymptomatic) submitted to elective castration were selected. The cohort size was determined based on the number of dogs presented to the HOVET-UFMT annually, taking into account a T. caninum infection prevalence of 10.9% (Pinto et al., 2014), and error margin ≤ 0.05.
This study was approved by the Committee for Ethics in Animal Use (CEUA) of UFMT under nº 23108.102289/2015-92 and executed according to the ethical principles. Informed consent was signed by the owner of each dog.

Clinical analysis: dermatology and complementary examinations
Dermatological, epidemiological, and general clinical evaluations were performed on the dogs. The data related to the dermatological changes included macroscopic morphological characteristics (location, shape, and the intensity of changes) (Ginn et al., 1972;Scott et al., 2001). Based on the cutaneous changes found in the dogs, complementary tests including direct parasitological examination of samples acquired via skin scrapings or adhesive tape (Pereira et al., 2015); wood lamp examination; direct microscopic hair examination, after imbibition in 10% potassium hydroxide, fungal culture, and cytological examination of fine-needle aspirates of skin samples (Bond, 2010) were performed for a definitive diagnosis of the dermatopathy.

Detection of Trypanosoma caninum and Leishmania infantum infection
Blood samples (5 mL) were collected by cephalic or jugular venipuncture using the aseptic technique into 10% sodium EDTA tubes for parasitological culture and molecular tests. In addition, five skin biopsies of dermatologically asymptomatic dogs and five skin biopsies from cutaneous lesions of dermatologically symptomatic dogs were collected. Three-millimetre tissue biopsies were obtained after trichotomy, antisepsis, and 2% lidocaine local anaesthesia (Almeida et al., 2010) for parasitological culture, molecular testing, and histopathology.

Parasitic isolation in culture medium
Skin biopsies were immersed in sterile buffered saline containing penicillin, streptomycin, and fluorocytosine, and preserved at 4 °C for 24 hours. The samples were then seeded into screw-cap tubes containing blood-agar slants biphasic medium (Novy, MacNeal, Nicole -NNN -Sigma-Aldrich ® ) overlaid with 2 mL of Schneider's Drosophila medium (Sigma-Aldrich ® ) plus 10% fetal bovine serum (Sigma-Aldrich ® ). Peripheral blood collected in tubes containing anticoagulants were seeded directly into the culture medium (400 μL). The cultures were seeded in duplicates, placed in an incubator at 26-28 °C, and examined weekly for 40-50 days by fresh tests to detect flagellate forms (Almeida et al., 2011).
DNA extraction, Polymerase Chain Reaction (PCR) and sequencing DNA extraction from the skin and blood samples were performed using the phenol/chloroform method (Sambrook & Russell, 2001). For the detection of T. caninum DNA, nested PCR was used with the outer primers TRY927F and TRY927R during the first round and the inner primers SSU561F and SSU561R, which amplify a conserved region of the 18S rDNA (ribosomal DNA) gene of all trypanosomatids (Smith et al., 2008), during the second round. For the detection of an L. infantum infection, RV1 and RV2 primers were used to amplify a 145 bp region of kDNA (kinetoplast DNA) specific for this species (Lachaud et al., 2002). For both sets of primers, the L. infantum DNA (MHOM/BR/1974/PP75) was used as the positive control and a corresponding negative control (DNA free reaction) was included in every reaction.
The amplification products of the nested PCR and L. infantum specific PCR were fractionated by 2% agarose gel electrophoresis, stained with Red gel (Biotium ® ) and viewed with ChemiDoc XRS using the Image Lab Software. A 100 bp molecular weight marker (Fermentas ® ) was used. The products obtained by the 18S rDNA nested PCR were purified using the GFX PCR DNA kit and sequenced using Sanger sequencing (Applies Biosystems ® Genetic Analysis) for etiological characterization. The sequences obtained were deposited and compared to sequences of different species of trypanosomatids using GenBank.

Statistical analysis
The data were tabulated and analysed descriptively. The association of sex, breed and age with the presence of dermatological changes and T. caninum or L. infantum infection was examined using chi-square or Fisher Exact tests, using statistical software (Epi Info 3.3.2 program, CDC, Atlanta, GA). A P-value < 0.05 was considered significant.
In the parasitological culture, flagellar forms were isolated from two skin biopsies from two different dogs. By molecular analysis, trypanosomatid DNA was detected in 18 (12%) dogs, independent of the samples and primers used ( T. caninum and L. infantum infections were not significantly associated (p>0.05) with the breed, sex, age, or the presence of dermatopathies. T. caninum infection was detected in two (1.3%) asymptomatic dogs and one (0.7%) dog which presented skin pruritus with alopecia and desquamation. L. infantum infection was found in three (9.1%) asymptomatic dogs and twelve (10.2%) dogs which showed the following clinical findings: alopecia (75%), desquamation (75%), pruritus (67%), cutaneous ulceration (33%), and onychogriphosis (25%).
One hundred of the fungal cultures (66.7%) were positive for Microsporum spp. or Trichophyton spp., of which, twelve dogs (12%) were positive for L. infantum and three (3.0%) were positive for T. caninum. Demodex or Sarcoptes mites were detected by skin scrapings or adhesive tape in eleven dogs (7.3%). One (0.7%) of the dogs detected with demodicosis was positive for L. infantum.
The skin histopathology results of the infected dogs are presented in Table 2. In three L. infantum infected dogs (20%), no histopathological abnormalities were observed. In one dog, the changes were suggestive of endocrinopathy, characterized by acanthosis, epidermal and follicular infundibulum orthokeratotic hyperkeratosis, and hair follicle and sebaceous gland atrophy.

Discussion
The present study conducted in an endemic area of T. caninum and L. infantum showed infection occurrence rates of 2% and 8% for these parasites, respectively, in agreement with previous findings Pinto et al., 2014). Infections by different protozoa of the Trypanosomatidae family in canines have been described previously (Nwoha, 2013). In this scenario, L. infantum infection has important public health implications (Madeira et al., 2006) due to the relevance of dogs as a reservoir of zoonotic visceral leishmaniasis. The recent isolation of T. caninum from dogs' skin samples in endemic areas for visceral leishmaniasis has called attention to this protozoan (Barros et al., 2012;Oliveira et al., 2015).
In addition, T. caninum infection was identified only in skin samples, even when parasitological culture and molecular analysis of blood samples were utilized (Pinto et al., 2014;Madeira et al., 2014). Skin was also the biological sample with the highest detection rate of L. infantum in symptomatic and asymptomatic dogs, independent of the target gene sequences used. Despite the low number of T. caninum positive dogs, the present results reinforce the finding that T. caninum infection is restricted to the skin . Conventional PCR and parasitological culture, presently utilized for identifying T. caninum infection, might not be sensitive enough to detect very low parasite burdens, possibly limiting the detection of this parasite, especially since little is known of its pathogenesis and biological cycle (Almeida et al., 2011;Madeira et al., 2014). However, the importance of dogs in canine visceral leishmaniasis (CVL) reinforces the need to promote the distinction of these agents (Barros et al., 2012).
Although the 18S rDNA gene detects Leishmania DNA, the use of the primers with a conserved region of L. infantum kDNA employed in this research obtained a higher positivity for this region, which makes it a good diagnostic method for visceral leishmaniasis, especially in endemic and sympatric areas. These results are consistent with those obtained by Pinto et al. (2014). However, regarding L. infantum, it is important to submit several biological samples for specific PCR and parasitological cultures, in order to have higher chances of detecting this agent (Almeida et al., 2011. The use of specific T. caninum primers might be more effective for its detection, as described for LVC. In this study, there was no association of T. caninum or L. infantum infection with age, breed, and sex, which is in agreement with previous findings in cases of visceral leishmaniasis (Almeida et al., 2009;Oliveira et al., 2010), while T. caninum has been detected in asymptomatic dogs aged 3 to 5 years (Oliveira et al., 2015). Nevertheless, the present T. caninum infection rate should be accepted cautiously, due to the low number of infected dogs.
Dermatological changes are frequently reported in L. infantum infections in canines (Queiroz et al., 2011;Silva et al., 2017). In this study, the main dermatological signs in dogs that were positive for this agent were alopecia, desquamation, pruritus, skin ulcers, and onychogriphosis. These findings are consistent with those of other authors (Almeida et al., 2009;Solano-Gallego et al., 2009). The occurrence of cutaneous coinfections, mainly fungal are the result of opportunist agents , but concomitant dermatological diseases can be associated with LVC. Nevertheless, in cases of LVC, many dogs may be asymptomatic (Almeida et al., 2009;Barros et al., 2012), as was observed herein.
The isolation and detection of T. caninum DNA exclusively in intact skin has been considered a peculiar aspect of this agent . According to Alves et al. (2012), T. caninum can provoke an insufficient mild humoral response to cause clinical lesions, however, no significant histopathological changes related to this infection have been described yet. In this study, only one dog positive for T. caninum infection presented dermatological abnormalities, characterized by multifocal alopecia, meliceric crusts, and generalized desquamation. Nevertheless, the present results should be interpreted cautiously, since this dog was also positive for dermatophytosis based on fungal culture (Mattei et al., 2014), which affects keratinized tissues (Nweze, 2011), and leads to similar morphological changes as described in the presented dog (Bond, 2010).
The main histopathological change seen in asymptomatic animals with T. caninum infection was a mild lymphoplasmacytic inflammatory infiltrate ( Figure 1A). However, the significance of this mild infiltrate might be questionable, because lymphocytes and plasma cells in small numbers are also present in the dermis of healthy dogs (Hargis & Ginn, 2007). The finding of acanthosis, epidermal orthokeratotic hyperkeratosis, and presence of melanomacrophages in the dermis of the dog with dermatological changes could not be associated with T. caninum infection due to the fungal co-infection with Microsporum sp. and Trichophyton sp., which produce similar changes (Bond, 2010). Histopathological changes consistent with dermatitis have been described in CVL (Queiroz et al., 2011). Perivascular and perifollicular changes are the most common, even in absence of the Leishmania sp. amastigotes ( Figure 1B). However, similar to the present study, chronic dermatitis, manifested by mononuclear cell, including plasma cells, macrophages and lymphocytes, infiltration was also described in dogs with and without CVL, and in those with the presence or absence of amastigotes (Figueiredo et al., 2010). The presence of mild lymphohistioplasmacytic inflammatory infiltrate that is commonly found in healthy tissue was also observed in dogs with leishmaniasis without clinical symptoms as well as in dogs infected with T. caninum. However, in animals with visceral leishmaniasis, an asymptomatic infection may persist for years before the development of clinical symptoms (Baneth et al., 2008); whereas, in a T. caninum infection, it is not yet known what the incubation period is or if this parasite is capable of producing clinical signs in infected dogs throughout the course of the disease. The presence of more than one inflammatory pattern, mixed inflammatory patterns, or an overlapping of histological patterns may be observed in L. infantum infections in dogs (Torres et al., 2008). CVL leads to diverse clinical and histological manifestations and the latter mostly demonstrates mild to marked inflammatory infiltrates, depending on host resistance variations (Calabrese et al., 2010). These findings may change due to cutaneous coinfections and concomitant diseases (Gross et al., 2005), such as dermatophytosis fungal, observed in 64.7% of L. infantum positive dogs herein, or hypothyroidism, where the characteristics were more suggestive of atrophic dermatoses than that of parasitic dermatoses.

Conclusion
The presence of nonspecific changes found in the histology of dogs with and without dermatopathies leads us to infer that T. caninum may not induce a cutaneous inflammatory reaction. However, more studies are necessary to better characterize the clinical and histological changes induced by this agent.