Trypanosomatid infections in captive wild mammals and potential vectors at the Brasilia Zoo, Federal District, Brazil

Abstract Background Conservation projects in zoos may involve translocation of captive animals, which may lead to pathogen spread. Neotropical mammals are important hosts of Trypanosoma cruzi and Leishmania spp. the etiological agents of Chagas disease and Leishmaniasis respectively. Studies of trypanosomatid‐infected mammals and vectors (triatomines and sandflies) in zoos are important for the establishment of surveillance and control measures. Objectives We investigated trypanosomatid infections in captive wild mammals, triatomines and sandflies at the Brasília Zoo. Methods We collected triatomines during active bimonthly surveys, sampled sandflies using light‐traps and obtained blood samples from 74 mammals between 2016 and 2017. We used quantitative PCR to detect trypanosomatids in vectors and mammals. Results We found a colony of 19 Panstrongylus megistus in the porcupine unit and detected T. cruzi infections in five bugs. We captured 17 sandflies of four species including Nyssomyia whitmani and Lutzomyia longipalpis, but no Leishmania infection was detected. qPCR detected 50 T. cruzi‐infected mammals belonging to 24 species and five groups of mammals (Carnivora, Cetartiodactyla, Perissodactyla, Pilosa and Primates); Leishmania DNA was detected in 23 mammals from 15 species, mainly carnivores. We detected trypanosomatid infections in 11 mammals born at the Brasília Zoo. Conclusions Our results suggest vector‐borne transmission of T. cruzi among maned wolves; measures to reduce the risk of new infections should therefore be taken. We also report sandfly presence and Leishmania‐infected mammals at the Brasília Zoo. Translocation of wild mammals in and out of the Brasília Zoo should consider the risk of T. cruzi and Leishmania spread.


| INTRODUC TI ON
Zoos play a key role in ex situ conservation by engaging visitors, community and employees (Barongi, Fisken, Parker, & Gusset, 2015).
Scientific research and species reproduction are also fundamental activities in zoos. If necessary, reintroduction programs for captive animals are developed to increase the free-living population and, in many cases, save species from extinction (Bowkett, 2009). Longterm population viability often requires transfering animals among institutions for breeding (Barongi et al., 2015). However, the risk of pathogen transmission by moving infected individuals should be evaluated. The spread of zoonoses is among the potential consequences of translocation. Infectious diseases are transmitted between hosts by a variety of mechanisms, including direct, aerial and vector-mediated transmission (Fèvre, Bronsvoort, Hamilton, & Cleaveland, 2006). Translocated animals can carry pathogens to the destination environment, acquire new ones during translocation or even become infected in this new environment (Leighton, 2002).
Phlebotomines were also reported in a Brazilian zoo, where high frequency of Nyssomyia whitmani was found (Teodoro et al., 1998).
Moreover, Souza et al. (2010) reported the infection of six crab-eating foxes (Cerdocyon thous) and a bush dog (Spheotos venaticus) by Leishmania infantum kept in the zoo of the Federal University of Mato Grosso. Leishmania infantum was also detected in C. thous, Puma concolor and Panthera onca at a zoo in the state of São Paulo (Da Silva Tenório et al.., 2011;Dahroug et al., 2010). After the death of a bush dog with Visceral Leishmaniasis at the Zoobotanical Foundation of Belo Horizonte, Minas Gerais State, 14 other canids kept in captivity at this institution were examined and four were positive for L.
In the Federal District of Brazil (FD hereafter), estimates indicate high rates of triatomine infection by T. cruzi (Minuzzi-Souza et al., 2018), which may be related to the enzootic transmission of this parasite in gallery forests. For example, 33% of 18 white-eared opossums (Didelphis albiventris) were infected by T. cruzi in the forest adjacent to the Brasília Zoo (Gurgel-Gonçalves et al., 2004).
Leishmania infantum-infected dogs have also been confirmed in FD (Cardoso et al., 2015), where sandflies are frequent in gallery forests (Ferreira, Minuzzi-Souza, et al., 2015;Rapello et al., 2018) and houses (Carvalho, Bredt, Meneghin, & Oliveira, 2010). Studies of trypanosomatid-infected mammals and vectors (triatomines and sandflies) in zoos are important for the establishment of surveillance and control measures for leishmaniasis and Chagas disease, including valuable information for animal health and management strategies such as translocations. Therefore, our aims were: 1) to identify triatomines and sandflies and their infection rates; 2) to determine the frequency of infection with T. cruzi and Leishmania spp. among captive wild mammals; and 3) to investigate vector-borne transmission of these trypanosomatids at the Brasília Zoo.

| MATERIAL AND ME THODS
The Brasília Zoo is in the Brazilian Federal District, Center-West Region of Brazil (−15.8475°S, −47.9392°W). The institution was established on 1957 in an area of 619.7 hectares and has approximately 900 animals. We distributed HP light traps (Pugedo, Barata, França-Silva, Silva, & Dias, 2005) in 22 points ( Figure 1): 11 in the gallery forest and 11 in the zoo's units (primates, lowland tapir, crab-eating fox, maned wolf, giant otter, giant anteater, puma, jaguar and Brazilian porcupine). At each point, we set one HP light trap daily at 5:00 p.m. and collected the trap on the following day at 7:00 a.m. for four consecutive days, bimonthly, between 2016 and 2017. In the last month, we used Shannon traps in all capture points where HP-traps were previously installed. We separated sandflies by sex, date and location. We mounted whole males and female heads and distal abdomen segments following Forattini, Ferreira, Rocha Silva, and Rabello (1973); we then identified all specimens to species level using the keys by Galati (2017). We stored female thoraces and proximal abdominal segments in PBS 1× at −20°C for DNA extraction.
We sampled triatomines in mammal units bimonthly for four consecutive days, between September 2016 and September 2017.
In addition, there was daily surveillance work by zookeepers who were instructed to catch insects morphologically similar to triatomines. We identified triatomines using dichotomic keys (Lent & Wygodzinsky, 1979). We examined the faeces of the collected triatomines by optical microscopy according to Minuzzi-Souza et al. (2018); fresh (400x) and Giemsa 10%-stained (1000x) hindgut contents were examined to detect T. cruzi (Cuba Cuba, 1998).
We sampled blood from wild captive mammals at the Brasília Zoo between 2016 and 2017. We did not sample individuals that were difficult to contain, such as cervids, individuals debilitated due to health problems, pregnancy or breastfeeding, and old-age animals.
We collected blood after physical, chemical or physicochemical containment. The volume of blood obtained varied according to the individual's weight and their physiological and clinical characteristics.
Most collections took advantage of routine containments following the Zoo's veterinarian timeline and emergency restraints, minimizing stress for the animals. Part of the collected blood (300 μl) was impregnated on filter paper, which was used for DNA extractions. We stored all collected material at −20°C.

| RE SULTS
In total, we surveyed 63 sites for triatomines between 2016 and 2017 at the Brasília Zoo and only one site was positive (1.5% infestation rate). We found a colony of Panstrongylus megistus in the Brazilian porcupine (Coendou prehensilis) unit, behind the nest box ( Figure 2) on September 2016. We captured 13 females, four males, two nymphs and 32 eggs of P. megistus. Of the 16 triatomines examined, one was found infected with T. cruzi by microscopy (trypomastigotes and epimastigotes) and four were positive by nDNA-qPCR. T.
cruzi DNA was not detected in the porcupine blood sample.
We collected blood samples from 74 mammals, belonging to six orders, 15 families and 32 species (Table 1) (Table 1 and Data S1).
We detected Leishmania DNA in 23 individuals from 15 species (Figure 3), resulting in an infection rate of 31.1%. Infected individuals belonged to the order Carnivora (n = 14), Pilosa (n = 5) and Primates (n = 4); only two individuals were born at the Brasília Zoo (Table 1 and Data S1).

| D ISCUSS I ON
Our study identified triatomines and sandflies at the Brasília Zoo and revealed a high frequency of infection with T. cruzi and Leishmania spp. among captive wild mammals, mainly carnivores. Some of the mammals were born at the zoo and we suggest vector-borne transmission. Here, we discuss ecological, epidemiological and veterinary implications of these results.
We detected the second colony of P. megistus at the Brasília Zoo. The first colony was detected in the small-primate unit, where vector-borne transmission of T. cruzi was confirmed (Minuzzi-Souza et al., 2016). Panstrongylus megistus was also detected at the Rio de Janeiro Primatology Center (Lisboa et al., 2004), where four individuals were captured and T. cruzi transmission was proven. The presence of infected P. megistus specimens is a potential risk factor for T. and may include insect in their diets) or even on infected free-living mammals (e.g. rodents, marsupials) that inhabit the area of the institution. Vertical transmisson is also possible, but our data do not support infection from mother to offpring (see Data S1). Moreover, our previous study analysing primate samples (Minuzzi-Souza et al., 2016) showed that three primates born to qPCR-negative mothers at Brasília Zoo were infected with the same T. cruzi strain as P. megistus caught in their lodgings. This finding is strongly suggestive of vector-borne transmission instead of vertical transmission, which is rare among wild primates (Lisboa et al., 2015). Moreover, low phlebotomine density could be explained by seasonal variation; e.g., the abundance of Lu. longipalpis in Mato Grosso do Sul was higher in the wet season of the second year of sampling (Oliveira, Galati, Fernandes, Dorval, & Brazil, 2008). In addition, all methods for vector detection are imperfect; future sampling should consider other detection methods and analytical approaches (Padilla-Torres, Ferraz, Luz, Zamora-Perea, & Abad-Franch, 2013;Valença-Barbosa, Lima, Sarquis, Bezerra, & Abad-Franch, 2014). We did not capture phlebotomines in the gallery forest adjacent to the zoo. We expected to capture sandflies in this area because they are frequent in gallery forests of the FD (Ferreira, Macedo, et al., 2015;Rapello et al., 2018). The absence of Leishmania-infected sandflies may be directly related to the low number of specimens captured and the low rate of infection by trypanosomatids in the studied area (Ferreira, Minuzzi-Souza, et al., 2015).
The rate of T. cruzi-infected primates was similar that previously described by Minuzzi-Souza et al. (2016). We found infected species that had not been studied in the previous work (Saguinus bicolor, Cebus albifrons, Alouatta caraya, Ateles marginatus, Lagothrix cana, Chiropotes satanas and Callicebus cupreus), whereas Aotus nigriceps, L. chrysomelas and Saguinus niger had already been found infected at the Brasília Zoo. To the best of our knowledge, the present study is the first to identify T. cruzi in L. cana, A. marginatus and C. satanas. We also detected canids and felids infected by T. cruzi. Zetun, Lucheis, Troncarelli, and Langoni (2014) identified T. cruzi-infected crab-eating foxes (C. thous) at the Sorocaba zoo, state of São Paulo. Ocelots (L. pardalis), maned wolves (C. brachyurus) and crab-eating foxes were positive for T. cruzi in an area close to the Serra da Canastra National Park (Rocha et al., 2013). Albuquerque and Barreto (1968) identified T. cruzi in a hoary fox (Lycalopex vetulus) captured in the region of Franca, São Paulo State, and our study is possibly the first to confirm the infection of a bush dog (Speothos venaticus), a little-known and endangered canid. Coatis (Nasua nasua), Neotropical otters (Lontra longicaudis), a crab-eating raccoon (Procyon cancrivorus) and an Andean bear (Tremarctos ornatus) were also infected by T. cruzi. This parasite was isolated from coatis in the Pantanal region (Herrera et al., 2008) and from a raccoon captured in the state of São Paulo (Barretto & Ferriolli Filho, 1970) (Monteiro et al., 2006;Monteiro, Dietz, & Jansen, 2010). After an 11-year follow-up research with L. rosalia and L. chrysomelas, Lisboa et al. (2015) concluded that these species maintain the infection for a long period and  (Malta et al., 2010); a bush dog and a hoary fox had also symptomatic Visceral Leishmaniasis in this zoo (Luppi et al., 2008). We recommend that veterinarians working with such species pay attention to signals of leishmaniasis and Chagas disease. Future studies should reveal the real impact of Leishmania spp. and T. cruzi infection in wild mammals. Desenvolvimento Científico e Tecnológico -CNPq.

CO N FLI C T O F I NTE R E S T
None.

E TH I C A L S TATEM ENT
The authors confirm that the ethical policies of the journal, as noted on the journal's author guidelines page, have been adhered to and the appropriate ethical review committee approval has been received. The US National Research Council's guidelines for the Care and Use of Laboratory Animals were followed.