Serological Evidence of Infection by Leishmania (Leishmania) infantum (Synonym: Leishmania (Leishmania) chagasi ) in Free-Ranging Wild Mammals in a Nonendemic Region of the State of São Paulo,Brazil

Introduction

Visceral leishmaniasis (VL) is one of the six endemic diseases considered priorities in the world because of its high incidence and mortality rates. The etiologic agent in the New World is Leishmania (Leishmania) infantum (synonym: L. (L.) chagasi) (Kinetoplastida: Trypanosomatina: Trypanosomatidae) (BRASIL 2006), and the vector species most involved in the transmission of American visceral leishmaniasis (AVL), the typical form in the Americas, is Lutzomyia longipalpis (Diptera: Psychodidae: Phlebotominae) (Lutz and Neiva 1912), although Lu. cruzi has also been implicated as a vector (FUNASA 2002, World Health Organization 2010).

Notwithstanding the lack of consensus about the taxonomic classification of L. (L.) chagasi (Cunha and Chagas 1937), which many authors consider a distinct species of L. (L.) infantum and responsible for AVL, evidence shows that they are, in fact, the same species (World Health Organization 2010, Kuhls et al. 2011).

Like humans, domestic and wild mammals are also susceptible to AVL, but dogs (Canis familiaris) play a major role in the domestic transmission cycle, acting as reservoirs. In the sylvatic cycle, the reservoirs are the canids Lycalopex vetulus (hoary fox) and Cerdocyon thous (crab-eating fox) and the didelphid Didelphis albiventris (white-eared opossum) (BRASIL 2006, World Health Organization 2010). Natural or anthropogenic environmental changes have facilitated the contact between these and other wild species with domestic animals and humans. Thus, increasing concern has focused on the interface between wildlife, domestic animals, and humans in the transmission of VL (Curi et al. 2006).

In Brazil, the eco-epidemiology of AVL has changed since the 1970s, with a clear process of urbanization, expansion of endemic areas, and the emergence of new outbreaks of the disease, owing to extensive socio-environmental changes. Investigations into the role of wild mammals in this context are very important, including the assessment of exposure of these animals to the vector and the parasite, which can be done by means of serological tests (FUNASA 2002, BRASIL 2006).

The purpose of this study was to investigate the presence of anti–L. (L.) infantum antibodies through the direct agglutination test (DAT), in 528 free-ranging wild mammals of 38 species from the municipality of Botucatu, state of São Paulo, Brazil, which is considered “silent” for AVL in view of the absence of records of human or canine autochthony (Cutolo et al. 2013).

Materials and Methods

Results

Nine of the 528 mammals were positive by DAT, with anti–L. infantum antibody titers above the cutoff point of 1:320, making a total prevalence rate of 1.7% (9/528; 95% confidence interval [CI] 0.6–2.8%). Most of the seropositive animals showed a titer of 320 (4/9; 44.4%), but elevated titers of 1280, 2560, and 5120 were observed in one coati (Nasua nasua), one tayra (Eira barbara), and one lesser grison (Galictis cuja), respectively (Table 1). The prevalence rates for different orders, families and species of this study are described in Table 2. In addition, 58 mammals (58/528; 10.9%) were detected with antibody titers below the cutoff point adopted (Table 3).

Discussion

Most studies of L. (L.) infantum infection in wildlife species involve canids and opossums, because they are known reservoirs of the parasite in the sylvatic cycle in different regions. In this study, however, serologic evidence of infection by the parasite was recorded in scantily studied species, whose exact role in the epidemiology of VL remains little known.

Of the six different mammal species that were seropositive by DAT, i.e., Callithrix jacchus, Lepus europaeus, Sphiggurus villosus, E. barbara, N. nasua, and G. cuja, natural infection by L. (L.) infantum has only been recently confirmed in hares (Le. europaeus) in Spain (Ruiz-Fons et al. 2013). In Venezuela, animals of the Order Lagomorpha seem to be of particular epidemiological interest because of their wide geographical distribution, which coincides with leishmaniasis foci (Reyes and Arrivillaga 2009). In Brazil, there are previous records of the detection of anti–Leishmania spp. antibodies only in coatis (N. nasua), with a prevalence rate of 50.0% (1/2), also using the DAT, with a cutoff point of 1:10 (Voltarelli et al. 2009). The prevalence rate of 4.6% observed in this study for this species would increase to 39.5% (17/43) adopting similar cutoff point of 1:20.

Regarding the other seropositive species, little is known about the infection by parasites of the genus Leishmania. The white-tufted-ear marmoset (C. jacchus) has been used as an animal model in studies involving different species of Leishmania spp. (Marsden et al. 1981, Dietze et al. 1985). However, according to Carneiro et al. (2012), they are considered inadequate biological models for the study of VL, because they are able to inhibit experimental infections by L. (L.) infantum. The orange-spined hairy dwarf porcupine (S. villosus) belongs to the Family Erethizontidae, the same family of Rothschild's porcupine (Coendou rothschildi), species to which natural infections by another Leishmania species, L. (L.) hertigi (Herrer 1971), have been attributed, including in Brazil (Silva et al. 2013).

No canid or opossum was seropositive, although these species are the ones most frequently reported with infection by L. (L.) infantum, and despite the fact that D. albiventris was the species with the largest number of samples. Similarly, although bats constituted a large percentage of the animals sampled in this study and although they are considered one of the preferred food sources of Lu. longipalpis (Dias et al. 2003), no chiropteran presented detectable antibody levels.

Nevertheless, antibody titers lower than 320 were detected in the sera from 18 opossums, all five crab-eating foxes (Cerdocyon thous) and two of the eight hoary foxes (Lycalopex vetulus). Studies with opossums and canids in Brazil used the DAT with lower cutoff points, recording seroprevalence rates of 8.1% (9/111) for opossums, with titers of ≥40 (Schallig et al. 2007) and 35.0% (12/39) with a cutoff point of 1:10 (Voltarelli et al. 2009). In the latter study, 100.0% (2/2) of C. thous and L. vetulus were also positive, although the results of culture were negative for these animals. In this study, if a similar cutoff point of 1:20 or 1:40 was adopted, the prevalence rates for the opossums would be 6.2% (18/289) and 1.4% (4/289), respectively, and 100.0% (5/5) for C. thous and 25.0% (2/8) for L. vetulus with a titer of ≥20.

On the other hand, using the same cutoff point (1:320) adopted in our study for the DAT, Semião-Santos et al. (1996) reported a prevalence rate of 60.0% (3/5) for foxes in Portugal, albeit with confirmation of infection by parasitological examination or isolation. It is noteworthy, therefore, that the use of high cutoff points decreases the sensitivity and enhancing specificity of the test, leading to a lower prevalence, but with an equally lower number of false-positive results (Sundar et al. 2006). However, comparisons of prevalence should be cautious, because numerous differences may occur between studies.

In addition to canids and opossums, antibody titers also below the adopted cutoff point were detected in other species that were not considered seropositive, i.e., Myocastor coypus, Lutreolina crassicaudata, Myrmecophaga tridactyla, Tamandua tetradactyla, and Dasypus novemcinctus. The last two species have been associated with infections by parasites of the genus Leishmania in Brazil (Lainson et al. 1979, 1981, Araújo et al. 2012). In Venezuela, the nine-banded armadillo (D. novemcinctus) was related to infection by the parasite (Reyes and Arrivillaga 2009), and in Uzbekistan, the nutria (M. coypus) was described with clinical cutaneous leishmaniasis (Uralov 1980).

It is noteworthy that the possibility of infection in these animals with titers lower than 320 should not be discarded, because there is no standard cutoff point for the use of the DAT in different wild species. However, the test is frequently used in epidemiological studies involving humans and animals, and the technique was validated by several of these studies. Semião-Santos et al. (1995) reported a seropositivy agreement of 94.04% between DAT and IFAT/enzyme-linked immunosorbent assay (ELISA) in the dog reservoir, and De Korte et al. (1990) also reported comparable results between DAT and IFAT, ELISA, cross-over electrophoresis, and the latex agglutination test also in dogs, in addition to high specificity and absence of cross-reactions in humans patients. In addition, the DAT has been considered a reference in epidemiological studies (Sousa et al. 2011), and the technique is particularly important for seroepidemiological studies in wild species, because it has the advantage of not requiring species-specific anti-immunoglobulins, which is a limitation to the use of other serology techniques.

It is known that the genera Trypanosoma and Leishmania are closely related (Meredith et al. 1995), so the possibility of serological cross-reactions cannot be discarded. Serological cross-reactions may also occur with other species of the genus Leishmania, although the antigen used in this study is recommended for the detection of anti-L. donovani complex antibodies. However, the DAT generally provides excellent diagnostic accuracy, with high sensitivity and specificity reported by numerous authors (Silva et al. 2005, Sundar et al. 2006, Ritmeijer et al. 2006, Rajasekariah et al. 2008, Sukmee et al. 2008, Mahmoudvand et al. 2011, Sousa et al. 2011).

It should also be noted that the DAT is indicated as a screening test for LV when low and less specific titers are detected, whereas high titers allow the use of the test for the final diagnosis of the disease (Veeken et al. 2003). Therefore, the high titers of antibodies detected in a lesser grison, a tayra, and a coati stand out in this study, suggesting strong evidence of infection by L. (L.) infantum. These three species belong to the Order Carnivora, which also includes L. vetulus and C. thous, two species that are known natural reservoirs of the parasite. It seems important to investigate the exact role of wild species that have been reported with L. (L.) infantum infections, including those described in the present study, since it is clear that the classic cycle of VL, dog–sandfly–human, is probably much more complex than previously thought and may involve a network of animal species (Navea-Pérez et al. 2015).

Some studies suggest the possibility of direct involvement of wild species in the transmission cycle of VL as reservoirs, especially rodents, which can be particularly important because of their large and numerous distributions. These animals had confirmed Leishmania infection in endemic areas for human leishmaniasis in Ethiopia (Kassahun et al. 2015) and for canine leishmaniasis in Spain (Navea-Pérez et al. 2015), by molecular techniques. Also in Brazil, infection by L. (L.) infantum was detected by PCR in spleen and skin samples of wild rodents coming from rural areas (Lima et al. 2013). Another study described that the rodent Thrichomys laurentius was able to maintain the parasite in his viscera and skin after 12 months of experimental infection, even without any clinical or histological changes (Roque et al. 2010). However, to confirm the role of reservoirs of the different wild species, more complex studies are required, involving isolation of the parasite in natural infections, experimental infections, and xenodiagnostic, to prove their infectiousness to the competent vectors (Lima et al. 2013, Millán et al. 2014, Kassahun et al. 2015).

Despite the small number of sampled animals of some species, which resulted in large confidence intervals, and the low prevalence rates described here, it is important to note that all the nine seropositive animals belong to the wildlife of Botucatu. Although this municipality is in the path of expansion of AVL in southeastern Brazil, it has no evidence or record of autochthonous human or canine transmission, or of the occurrence of the disease vector, and is therefore considered a “silent” municipality (Cutolo et al. 2013).

Current measures for the control of AVL in Brazil, based on the elimination of the canine reservoir and the application of insecticides, have not sufficed to reduce prevalence rates in endemic regions (FUNASA 2002), or to prevent the introduction of the disease into new areas. Moreover, the transmission cycle of zoonotic VL in the country is concentrated in areas where human dwellings may be located in the proximity to the sylvatic cycle (World Health Organization 2010). Also, in Europe the dog is considered the main reservoir of L. (L.) infantum in endemic areas, and the existence of wild reservoirs have been proposed as a possible cause of the lack of success in controlling VL (Millán et al. 2014). Therefore, this underscores the importance of investigating the infection in wildlife and of the inclusion of “silent” regions in surveillance, which represents a new approach to disease control and a strategy to make it more effective (BRASIL 2006).

Conclusions

We have described serological evidence of infection by L. (L.) infantum in nine free-ranging wild mammals of the species C. jacchus, L. europaeus, S. villosus, N. nasua, E. barbara, and G. cuja, with a prevalence rate of 1.7% (9/528), in the municipality of Botucatu, São Paulo, Brazil, which has no records of autochthonous canine or human cases of VL. These findings highlight the importance of elucidation of the exact role of these species in the maintenance and transmission of the parasite, and the need for comprehensive and effective epidemiological surveillance in areas where the canine or human disease does not occur, especially those with increasing contact between urban and wild environments, to prevent its expansion.