Evaluation of canine leishmaniasis and concomitant seropositivity for Babesia canis and rickettsia in a nonendemic area in the central west region of Minas Gerais

Copyright Gama-Melo et al. This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License which permits unrestricted non-commercial use, distribution, and reproduction in any medium provided the original work is properly cited. Evaluation of canine leishmaniasis and concomitant seropositivity for Babesia canis and rickettsia in a nonendemic area in the central west region of Minas Gerais


Introduction
Leishmaniasis are part of a group of diseases with different clinical and epidemiological characteristics, which are considered neglected by the World Health Organization (World Health Organization, 2018). Visceral leishmaniasis (VL), also known as kala-azar, has viscerotropic characteristics and is considered a systemic zoonosis of chronic evolution, which if untreated, can be fatal (Brasil, 2014a). VL has been reported in five Brazilian regions, and the number of human VL (HVL) cases has increased significantly in Minas Gerais.
The presence of Lutzomyia longipalpis phlebotomine sandflies was detected in the municipality of Carmo da Mata, located in the central west region of Minas Gerais, but there are no reports of HVL cases and no epidemiological studies on canine VL (CVL) (Minas Gerais, 2016). According to the Visceral Leishmaniasis Control Program Manual (PC-VL) of the Brazilian Ministry of Health (MS), these characteristics define the city as a silent and receptive area for VL transmission (Brasil, 2014b).
Dogs are the main reservoir of Leishmania infantum parasites in urban environments, and canine cases of leishmaniasis precede human cases. Control strategies proposed by the MS include canine reservoirs, insecticide application, and diagnosis and early treatment of human cases. These actions should be executed together for better effectiveness (Brasil, 2014b).
Other canine diseases transmitted by arthropods, mainly Rhipicephalus sanguineus ticks, such as canine granulocytic anaplasmosis, babesiosis, and ehrlichiosis, have clinical signs similar to those of CVL (Dumler et al., 2001;Gonçalves et al., 2014). Recent studies have shown the occurrence of coinfection in dogs, with Leishmania and other parasites such as Babesia canis and rickettsia such as Anaplasma phagocytophilum and Ehrlichia canis in several regions of Brazil (Oliveira et al., 2008;Paulan et al., 2013;Gonçalves et al., 2014).
The close interaction between humans and dogs and the susceptibility of both species to vector-borne diseases make it relevant to conduct studies and evaluate the occurrence of these coinfections.
The main objective of the present study was to evaluate the existence and prevalence of CVL dogs seropositive for Babesia canis and rickettsia in the municipality of Carmo da Mata, Minas Gerais.

Material and methods
The study was approved by the Animal Ethics Committee of the Federal University of São João del Rei (protocol no. 016/2015). This cross-sectional study was conducted in the municipality of Carmo da Mata, Minas Gerais (Figure 1), which has an estimated total population of 10,927 inhabitants (Instituto Brasileiro de Geografia e Estatística, 2010). The study used the stratified sampling method to verify the existence of CVL and concomitant seropositivity for Babesia canis and rickettsia in the urban area of Carmo da Mata, between July and October 2015.

3/9
Evaluation of canine leishmaniasis and concomitant seropositivity for Babesia canis and rickettsia in a nonendemic area in the central west region of Minas Gerais The current study utilized the estimate of one dog for every five inhabitants (Brasil, 2014b), with 1,713 dogs estimated for an urban human population of 8,565 inhabitants.
According to the PC-VL of the MS, a canine serological survey should be performed in a municipality with silent VL transmission which has a canine population of more than 500 animals, and no previous studies on the prevalence of CVL (Brasil, 2014b).
The stratified sample of 430 dogs was calculated based on Table 1 of the PC-VL, estimating the canine population at over 1,000 dogs and a mean prevalence rate of 1.1 to 2% (used when there are no previous data on the existence of this parasitosis in the study area), and taking 5% as the level of significance (Brasil, 2014b).
Conglomerate sampling was used in the urban area, in which strata are sectors (neighborhoods) and conglomerates are blocks. These blocks were chosen using a random number table. For better spatial distribution of the sample in the strata, 50% of the households in the blocks were systematically included and canine blood collection started from the northernmost corner (Brasil, 2014b).
The cephalic or jugular vein was used to collect 5 mL blood samples from the 433 sampled dogs domiciled in the municipal urban area. Serum obtained after centrifugation at 3000 rpm for 10 minutes was frozen and subsequently tested for the evaluated parasites.
According to the MS protocol, only animals that are CVL-positive in the rapid immunochromatographic test (TR-DPP  ) should have a venous blood sample collected for ELISA (confirmatory). Animals that are positive for both diagnostic techniques are considered CVL-positive. In this study, all sampled animals underwent the two immunological tests for CVL diagnosis (TR-DPP  and ELISA). The existence of concomitant seropositivity was analyzed through indirect immunofluorescence reactions (IIFRs) for anti-Babesia, anti-Anaplasma, and anti-Ehrlichia antibodies.
The TR-DPP  and ELISA tests were performed using kits produced by Biomanguinhos/FIOCRUZ and following the manufacturer's recommendations. The tests were conducted at the Parasitology Laboratory of the Federal University of São João del Rei -Central-West Campus, which is accredited by the Ezequiel Dias Foundation as a regional reference center for CVL diagnosis. The IIFRs to diagnose babesiosis, anaplasmosis, and ehrlichiosis were conducted at the Veterinary Protozoology Laboratory of the Department of Parasitology of the Biological Sciences Institute (ICB) of the Federal University of Minas Gerais (UFMG).
E. canis and A. phagocytophilum antigens were obtained from infected IDE8 cells, using the method previously described by Aguiar et al. (2007). Babesia antigens were obtained from canine blood smears containing B. vogeli parasitized erythrocytes, prepared according to the Instituto Interamericano de Cooperación para la Agricultura (1987). For antibody screening, the slides were examined on an Olympus  BX41 epifluorescence microscope (Olympus Corporation, Tokyo, Japan) under 20X and 40X magnification. Fluorescent serological samples were diluted to 1:40.
A total of 371 dogs (85.7%) had no clinical signs suggestive of parasitic infection by Leishmania. Of the 62 remaining animals (14.3%) that presented some clinical signs, two presented all signs evaluated (adenopathy, alopecia, keratoconjunctivitis and/or blindness, dermatitis, foot edema, weight loss, onychogryphosis, and ulcers), nine presented two clinical signs concomitantly, four presented three concomitant signs, and 48 (11.1%) presented only one sign evaluated.

CVL diagnosis by TR-DPP  and ELISA
Of the 433 domiciled dogs evaluated in the urban area, 10 (2.3%) were positive in the TR-DPP  and, of these, three (0.7%) were positive in the ELISA. According to the PC-VL recommendations, only the dogs considered positive in the TR-DPP  and ELISA were considered positive for CVL. Therefore, the CVL prevalence for the domiciled dogs sampled in the urban area of Carmo da Mata was 0.7% (95%CI: 0.2-1.9).
The remaining 192 dogs were concomitantly seropositive for at least two of the evaluated pathogens.

5/9
Evaluation of canine leishmaniasis and concomitant seropositivity for Babesia canis and rickettsia in a nonendemic area in the central west region of Minas Gerais

Discussion
Dogs are important links in the transmission of the VL etiological agent and some studies have been conducted in the central west region of the state of Minas Gerais in order to better explain parasite dispersion in this region (Silva et al., 2008;Teixeira-Neto et al., 2014;Menezes et al., 2015;Faria et al., 2017).
There were no significant CVL differences in Carmo da Mata for sex and age of the animals, similar to the studies conducted in the western region of the state of Minas Gerais (Almeida et al., 2009;Santis et al., 2013;Faria et al., 2017). Clinical signs seen in CVL dogs in Carmo da Mata were similar to clinical signs observed in other studies and included skin changes, onychogryphosis, and weight loss (Almeida et al., 2012;Barata et al., 2013;Faria et al., 2017).
Some researchers stated that the serological tests used to diagnose CVL could present cross-reactivity with other parasites and rickettsia (Mancianti et al., 1996;Gomes & Cordeiro, 2004;Ferreira et al., 2007;Silva et al., 2011). However, Oliveira et al. (2008) rejected the hypothesis of cross-reaction among B. canis, E. canis, and Leishmania agents; they believe there are coinfections among these agents, especially in endemic regions. Due to biological differences, taxonomic classifications, and phylogenetic differences between rickettsia and protozoa, the hypothesis of coinfection is more acceptable than the occurrence of serological cross-reaction in the diagnosis of these infections (Dumler et al., 2001;Almosny et al., 2002;Krawczak et al., 2015.) Coinfection is a rarely studied phenomenon and most studies evaluate infections caused by only one agent (Sousa et al., 2013;Falcão et al., 2014). Most studies on the evaluation of multiple-agent infections in dogs are based on clinical reports (Gal et al., 2007;Sasanelli et al., 2009;Gonçalves et al., 2014;Silveira et al., 2015) which highlight severe clinical changes, difficult diagnosis, compromised treatment, and unfavorable prognosis (Vargas-Hernández et al., 2012).
In this study, the analyzed animals were evaluated for some agents transmitted by ticks to dogs and possibly to humans. Although tick infestation was not evaluated during collection, it was possible to see that some animals were infested with significant numbers of ticks. The large number of dogs concomitantly seropositive for tick-borne agents leads us to believe that the evaluated animals were infested with these arthropods. R. sanguineus ticks are frequently found in urban areas (Labruna & Pereira, 2001), besides being a species commonly found on dogs in Brazil (Oliveira et al., 2008;Morais et al., 2013).
It is hypothesized that R. sanguineus can act as mechanical CVL vectors because CVL and HVL have already been reported in areas free of phlebotomine sandflies (Michalsky et al., 2009). Moreover, L. infantum DNA was detected in ticks through PCR (Coutinho et al., 2005;Dantas-Torres, 2011;Colombo et al., 2011;Morais et al., 2013;Campos & Costa, 2014). However, a study by Paz et al. (2010) concluded that parasite maintenance and multiplication of Leishmania was unlikely in ticks. A recent study detected promastigote forms in the intestines, ovaries, and salivary glands of ticks, emphasizing the need for more studies on the role of ticks in CVL transmission in Brazil (Viol et al., 2016).
Tick-infested dogs present a 53% higher chance of being infected by L. infantum compared to non-infested dogs (Paz et al., 2010) because tick infestation causes anemia and weakness, increasing the chance of coinfection (Sousa et al., 2013).
There was no significant difference between the prevalence of B. canis in dogs in Carmo da Mata and the prevalence reported in other studies, such as a study in the city of São Paulo, where the prevalence was 42.4% (Dell'Porto et al., 1993).
The prevalence of canine E. canis rickettsia in Carmo da Mata was significantly higher when compared to prevalence rates reported in studies conducted in other Brazilian municipalities, such as 37.9% in Monte Negro, Rondônia (Aguiar et al., 2007); 36% in the microregion of Ilhéus and Itabuna, in Bahia (Carlos et al., 2007);and 42.5% in Cuiabá, in Mato Grosso (Silva et al., 2010). There was no significant difference between the prevalence found in Carmo da Mata and Belo Horizonte (Oliveira et al., 2008).
E. canis prevalence rates higher than those observed in Carmo da Mata have also been reported. Some studies in São Paulo showed E. canis infection prevalence of 75.3% in Ilha Solteira (Paulan et al., 2013), 78% in Jaboticabal (Oliveira et al., 2008), and 66.5% in Presidente Prudente (D' Andrea et al., 2015). These studies report a significant difference when compared to the prevalence noted in Carmo da Mata.
There are few studies on A. phagocytophilum infection in Brazil . A study conducted by Santos et al. (2013) in Rio de Janeiro using qPCR showed that 6% of the dogs were infected. A study on E. canis and A. phagocytophilum coinfection in dogs conducted in Belo Horizonte showed dogs infected by A. phagocytophilum for the first time in Brazil using parasitic and molecular diagnostic methods . This parasitosis should be further studied, since dogs act as sentinels for human exposure, posing as an infection source for ticks that can transmit the agent to humans (Falcão et al., 2014).
In the present study, the prevalence of A. phagocytophilum was 5.5% for mono-parasitized dogs and 38.3% for mono-and poly-parasitized dogs. None of the published studies on this subject used a random sample with more dogs evaluated than the present study. The prevalence of 38.3% found in this study is significantly different when compared to the rate found in a study conducted in southern Germany with Bernese Mountain Dogs, with a relative risk of 1.3 (95%CI: 1.1-1.6) (Preyß-Jägeler et al., 2016). A study conducted in Brazil reported a 35% occurrence of canine A. phagocytophilum in Campos dos Goytacazes, in the state of Rio de Janeiro (Falcão et al., 2014).
It is important to consider the hypothesis that the presence of one agent can act as a facilitator for the onset of other infections, which can be aggravated (Sousa et al., 2013). The high prevalence of E. canis is can result in immunosuppression (Boozer & MacIntire, 2003), as observed in this study, wherein the coinfections with higher prevalence were those involving E. canis infection.
The CVL and canine ehrlichiosis epidemiologic studies overlap in many areas of the world, which is justified by similar activity periods of agent vectors (Mekuzas et al., 2009). This indicates the importance of dog surveillance in the municipality of Carmo da Mata, since E. canis infection can contribute to an increased prevalence of CVL (Mekuzas et al., 2009). In addition, Carmo da Mata has some risk factors, such as the presence of L. longipalpis vector, CVL positive dogs, and proximity to other areas where there is a well-established transmission cycle, including HVL cases.

Conclusions
The low prevalence of CVL in Carmo da Mata indicates a possible expansion of this parasitosis in the central west region of Minas Gerais, which is considered safe by the Ministry of Health and Minas Gerais State Health Department.
Preventive measures including control of tick infestations in dogs are necessary in order to avoid a possible increase in the number of cases of tick-borne infection. Tick control will also decrease the occurrence of human infections caused by A. phagocytophilum, since dogs act as sentinels for the transmission of this disease to humans.
Even with the low prevalence found in this study, CVL should be controlled because of the high rate of E. canis positivity in dogs; the immunosuppression caused by ehrlichiosis can lead to an increase in the number of CVL cases.
We suggest that veterinarians should consider the differential diagnosis for the evaluated agents, since the occurrence of concomitant seropositivity may lead to high parasite rates and to the permanence of enzootic diseases in the municipality.