Canine rangeliosis is a hemorrhagic and hemolytic disease transmitted by the ixodid tick Amblyomma aureolatum ( ^{SOARES et al., 2018} ). It is caused by the protozoan parasite R. vitalii belonging to the order Piroplasmida that infects erythrocytes, leukocytes and endothelial cells ( ^{LORETTI & BARROS, 2005} ). In Brazil, R. vitalii has been reported in domestic dogs in southern and southeastern regions ( ^{GOTTLIEB et al., 2016} ; ^{LEMOS et al., 2017} ). Few studies report the biochemical abnormalities observed in the serum of dogs infected with R. vitalii ( ^{COSTA et al., 2012} ). The present study aimed to report the molecular diagnosis of R. vitalii in dogs naturally infected from the metropolitan area of the city of Curitiba, state of Paraná, south Brazil, and to describe the clinical, hematological and biochemical findings of the infected dogs.

This study was approved by the Ethics Committee of the Federal University of Paraná (UFPR) (approval no. 035/2015). The clinical records of three male, mixed-breed dogs (dogs 1-3), presented between November 2015 and February 2016 to the Veterinary Hospital from the UFPR, were reviewed. Veterinarians performed the physical examination, and blood samples were collected (3 mL) by venipuncture into EDTA tubes, and into tubes without anticoagulant. A small amount of blood was sampled from the capillary beds of the ear tip for peripheral blood smears.

A Mindray BC 2800 Vet^® automatic hematology analyzer was used for total erythrocyte and leukocyte counts and for hemoglobin concentration measurement. The microhematocrit technique was used to obtain the hematocrit. Total plasma protein (PPT) concentration was determined by refractometry. Differential leukocyte count and screening for hemoparasites were performed on Romanovsky-stained blood smears examined by light microscopy. The tubes without anticoagulant were centrifuged at 5000 rpm for 5 minutes to obtain serum. Biochemical parameters were determined in serum using commercial kits (Labtest^®) in a Mindray BS-200^® automated biochemical analyzer. Hematology and biochemistry results were interpreted based on canine reference ranges ( ^{KANEKO et al., 1997} ; ^{MEINKOTH & CLINKENBEARD, 2000} ).

Genomic DNA for PCR was extracted from blood samples collected in EDTA tubes via a PureLink™ Genomic DNA Mini Kit (Invitrogen^®), according to the manufacturer's instructions. Primers used were BAB143-16 and BAB694-667 previously described ( ^{ALMEIDA et al., 2013} ), and a 551-bp fragment of the Babesia sp. 18S rRNA gene was obtained after PCR DNA amplification. PCR products were analyzed on a 1.5% agarose gel stained with GelRed ^® and examined under UV transillumination. PCR products were sequenced by Sanger method (ABI PRISM^® 310 Genetic Analyser, Applied Biosystems), and the sequences obtained were submitted to BLAST analysis and compared with those in the GenBank to confirm the identity of the detected sequences.

Table 1 presents information on the three sick dogs including signalment, history, and clinical signs. One dog (dog 1) died. The hematological abnormalities in these animals are summarized in Table 2 . The main laboratory abnormalities in all three dogs were severe anemia, shift of neutrophils to the regenerative left, presence of toxic neutrophils, and thrombocytopenia. Neutrophilia was observed in two dogs (dos 1 and 2). Polychromatophilic erythrocytes were not observed. Intraerythrocytic and extracellular forms of the piroplasm were found in the peripheral blood smears of all three patients ( Figure 1 ). Serum biochemistry results are shown in Table 3 .

Figure 1 Photomicrograph of peripheral blood smear of dog (dog 2) naturally infected with R. vitalii. Presence of extracellular piroplasms, oil immersion objective, 1000x. Romanovsky stain.  

Blood samples from all three dogs were positive for Babesia sp. by PCR ( ^{ALMEIDA et al., 2013} ). However, sequence analysis of the PCR products showed that these were R. vitalii . Sequences were deposited in GenBank (Accession numbers: MG027583.1, MG027584.1 and MG027585.1). The DNA sequences were compared with those from other piroplasms of dogs deposited in GenBank. Our sequences showed high degree of homology (100% for dogs 1 and 2, and 99% for dog 3) with those of R. vitalii reported in the State of Rio Grande do Sul, south Brazil (KT323930.1), in the State of Rio de Janeiro, southeast Brazil (KU710789.1), and in Uruguay (KP202861.1) ( Figure 2 ).

Figure 2 Distance tree (Neighbor-Joining method), conducted using MEGA software (Molecular Evolutionary Genetics For Review Only Analysis) version 6.0 ( ^{TAMURA et al., 2013} ). Hepatozoon canis was used as outgroup control.  

In this study, intraerythrocytic and extracellular forms of R. vitalii were seen in peripheral blood smears from all three dogs. R. vitalii is morphologically similar to B. vogeli on blood smears, but genetically distinct in molecular phylogeny ( ^{LEMOS et al., 2017} ). In the State of Paraná, south Brazil, only one study detected R. vitalii in a dog by molecular diagnosis ( ^{MONGRUEL et al., 2017} ).

In the present report, the clinical signs observed in all three dogs included apathy, anorexia, dehydration, pale or icteric mucous membranes, and hemorrhages. These findings corroborate previous studies ( ^{LORETTI & BARROS, 2005} ; ^{FIGHERA et al., 2010} ; ^{GOTTLIEB et al., 2016} ; ^{FREDO et al., 2017} ). One dog with jaundice (dog 1) succumbed to rangeliosis. Icterus is a common necropsy finding in dogs with rangeliosis ( ^{FIGHERA et al., 2010} ).

In canine rangeliosis, coagulation disorders affecting primary hemostasis cause a hemorrhagic disease ( ^{PAIM et al., 2012} ). Two dogs from our study (dogs 2 and 3) had hematochezia, which is a clinical sign that has been observed by other authors ( ^{FIGHERA et al., 2010} ; ^{FREDO et al., 2017} ). Hemorrhagic diathesis is a common and striking clinical presentation of cases of R. vitalii infection in dogs ( ^{LORETTI & BARROS, 2005} ; ^{PAIM et al., 2012} ). The tips of the ears from one dog (dog 1) were bleeding when he was presented for veterinary care. Severe, bilateral pinnal hemorrhage is a typical clinical feature of canine rangeliosis, and occurs in 14.3% of the cases according to previous studies ( ^{FIGHERA et al., 2010} ).

Anemia and thrombocytopenia are the most common hematological findings in cases of canine rangeliosis ( ^{FIGHERA et al., 2010} ; ^{PAIM et al., 2012} ). The absence of polychromatophilia in our study differs from previous studies where most dogs with rangeliosis presented regenerative anemia ( ^{FIGHERA et al., 2010} ). In our study, hematologic abnormalities included leukocytosis and neutrophilia in two dogs (dogs 1 and 2) and regenerative left shift in all three dogs. These changes in the leukogram are associated with inflammation as the inflammatory response accelerates the production and release of neutrophils from the bone marrow resulting in neutrophilia with the presence of immature neutrophils in the circulation. Increased neutropoiesis induced by inflammation results in the release of toxic neutrophils into the bloodstream ( ^{SCHULTZE, 2010} ). The findings of regenerative left shift in all three dogs and lymphocytosis in dog 2 are consistent with findings from a recently published study about natural cases of canine rangeliosis ( ^{FREDO et al., 2017} ). In the present study, all 3 dogs were thrombocytopenic, corroborating previous studies ( ^{LORETTI & BARROS, 2005} ; ^{FIGHERA et al., 2010} ; ^{FREDO et al., 2017} ).

Hemorrhage and anorexia are the most likely explanations for the hypoproteinemia in three dogs ( ^{TVEDTEN, 2010} ). We suggest that hypoalbuminemia in dogs 1 and 2 occurred due to albumin is a negative acute-phase inflammatory protein ( ^{CERON et al., 2005} ). Previous studies evaluating the biochemical parameters of dogs experimentally infected with R. vitalii didn’t find any changes in the creatinine and urea serum levels ( ^{COSTA et al., 2012} ). In our study, there were also no abnormalities in the creatinine dosages of all three dogs with rangeliosis. However, urea levels were increased in all three dogs, most probably due to anorexia and degradation of endogenous proteins ( ^{KANEKO et al., 1997} ). AST and GGT enzymes, and total and direct bilirubin levels were increased in dog 1. In this report, elevated AST levels indicate hepatic injury. Elevated GGT levels and increased serum direct bilirubin levels are indicators of cholestasis and hepatic jaundice ( ^{STOCKHAM & SCOTT, 2011} ).

The present study highlights the importance of the occurrence of the canine rangeliosis in the region and the need to include it as a differential diagnosis for other hemoparasites.