Out of the 20 recognized species of armadillos in the world, 11 are found in Brazil, and five of them are found in Pantanal, one of the world's largest wetlands. Beef cattle (Bos taurus) farming is the main economic activity in this region, which promotes intense wildlife-livestock contact and increases the likelihood of pathogen exposure, including to agents with zoonotic and economic relevance. Previous studies demonstrated that several wildlife species in Pantanal have been exposed to Brucella abortus and Leptospira spp.; however, little is known regarding the exposure and/or prevalence of zoonotic pathogens in armadillos. We used conventional PCR, the rose Bengal test (RBT), and the microscopic agglutination test (MAT) to investigate the exposure to and infection by Brucella spp. and Leptospira spp. using blood samples from four species of armadillos: nine-banded armadillo (Dasypus novemcinctus, n=2), southern naked-tailed armadillo (Cabassous unicinctus, n=8), yellow armadillo (Euphractus sexcinctus, n=16), and giant armadillo (Priodontes maximus, n=22), captured in Nhecolândia, Pantanal, Brazil. Samples were PCR- and RBT-negative for Brucella spp. infection and exposure. However, MAT revealed a Leptospira spp. seroprevalence of 31% (5/16; 95% confidence interval [CI]=0.11–0.58) in yellow armadillo and 18% (4/22; 95% CI=0.05–0.40) in giant armadillo specimens to serogroups Autumnalis, Cynopteri, and Pomona, with titers ranging from 200 to 1,600. Our results contribute to the understanding of zoonotic pathogens in armadillos in Pantanal and reinforce the importance of wildlife health surveillance in this area.
Twenty species of armadillos belong to the order Cingulata, and five of the 11 species that occur in Brazil have been recorded in the Pantanal (Tomas et al. 2010). Pantanal is one of the world's largest wetlands, and it supports many populations of endangered species, such as the giant armadillo (Priodontes maximus), due to its high level of habitat conservation (Trolle 2003).
The human–wildlife–domestic animal interface is constantly challenged in the Pantanal. Beef cattle (Bos taurus) farming, the main economic activity in the region, forces wildlife to share their habitat with livestock, particularly in the rainy season (Vieira et al. 2016). Additionally, armadillos are hunted in rural areas for human consumption and domestication (Capellão et al. 2015).
Brucellosis is a zoonosis that occurs in a wide range of species worldwide, it is transmitted through contact with reproductive tissues of infected animals or contaminated food, and clinical manifestations include infertility, abortions, orchitis, and arthritis (Thorne 2001). Leptospirosis transmission occurs directly, by contact with body fluids from infected animals, or indirectly, by exposure to an environment contaminated by urine of infected animals (Day et al. 1998). Clinical signs may be inapparent; however, depending on the serovar, they can be involved in serious disease (Picardeau 2013). In farm animals, the infection may cause reproductive failure and leads to economic losses (Adler and Moctezuma 2010).
The dynamics of diseases in wildlife are changing due to anthropogenic activities and climate change, and pathogen monitoring is urgently needed to further understand wildlife disease dynamics (Lindahl and Grace 2015). Previous studies demonstrated exposure to Brucella abortus and Leptospira spp. in several wildlife species in Pantanal (Vieira et al. 2016). In spite of that, information regarding zoonotic pathogens in armadillos from Pantanal is scarce. Considering the close contact of wildlife and beef cattle farming in Pantanal, the objective of this research was to perform serologic and molecular tests to determine the exposure to and occurrence of Brucella abortus and Leptospira spp. in armadillos, as well as to investigate if there is an overlap with the known Leptospira serogroups of cattle in the Pantanal.
Blood samples from 48 armadillos were collected between 2011 and 2017: nine-banded armadillo (Dasypus novemcinctus, n=2), southern naked-tailed armadillo (Cabassous unicinctus, n=8), yellow armadillo (Euphractus sexcinctus, n=16), and giant armadillo (n=22), all captured in several ranches (19°15′S, 55°47′W) in Nhecolândia, Pantanal, Brazil (Fig. 1). As part of a health monitoring program, armadillos were captured, anesthetized, sampled, and released (Souza 2016).
Serologic diagnosis of leptospirosis was performed using the microscopic agglutination test. Succinctly, the serum samples were tested against 21 live antigens (Table 1) grown in Ellinghausen-McCullough-Johnson-Harris medium (Becton-Dickson, Sparks, Maryland, USA). To a 96-well microplate, we added 50 µL of serum diluted 1:50 with pH 7.6 buffered Sorensen's saline solution and 50 µL of antigen suspension, mixed by agitation and incubated at 28 C for 2 h. The plates were examined using a darkfield microscope. Sera presenting agglutination at dilutions ≤50% in comparison to the positive control were considered positive. As a positive control, we used sera from Mesocricetus auratus inoculated with the antigens. Samples that reacted were retested in several twofold dilutions beginning at 1:50, until we reached the maximum positive dilution. Samples with titers ≤100 were considered positive. Cross-reactivity between different serogroups may occur, so we considered the highest microscopic agglutination test titer as the predominant serogroup. To check auto-agglutination, we used 25 µL of buffered saline solution as a negative control and added each antigen to it.
We used the rose Bengal test (RBT) to detect Brucella abortus antibodies. The RBT does not detect Brucella canis or Brucella ovis. We mixed 30 µL of serum samples with an equal volume of the antigen (8% Brucella abortus 1119-3 whole-cell suspension buffered in 3.65 pH, Instituto Biológico, São Paulo, Brazil) for 4 min. A sample was considered positive when visible agglutination was observed (Alton et al. 1988).
For PCR, DNA from whole blood or a clot was extracted using the phenol-chloroform method (Vieira 2004). The concentration and purity of DNA were determined by spectrophotometry with an Epoch microplate spectrophotometer (Biotek®, Winooski, Vermont, USA). Conventional PCR for Brucella spp. was performed with primers B4 and B5 to amplify a 223-base pair fragment of the bcsp31 gene encoding the 31-kDa cell surface protein of the genus Brucella (Baily et al. 1992). The PCR cycle conditions were adapted to initial denaturation at 94 C for 5 min, template denaturation at 94 C for 45 s, primer annealing at 60 C for 45 s, primer extension at 68 C for 30 s and 40 cycles, with a final extension at 68 C for 3 min, in a C1000 Touch Thermal Cycler (BioRad Laboratories, Hercules, California, USA).
Conventional PCR for Leptospira spp. was performed with primers LEP-1 and LEP-2, targeting the 16S rRNA region (Mérien et al. 1992), under the following PCR cycle conditions: initial denaturation at 94 C for 5 min, 30 s of template denaturation at 94 C, 30 s of primer annealing at 60 C, and 30 s of primer extension at 72 C for 40 cycles, with a final extension at 72 C for 5 min, in a GeneAmp® PCR System 2700 (Applied Biosystems, Foster City, California, USA).
Brucella abortus (vaccine RB-51) DNA and Leptospira borgpetersenii DNA were used as positive controls, and nuclease-free water was used as a negative control. The amplification products were read in standard 1.5% agarose electrophoresis using SYBR® safe DNA gel stain (Thermo Fisher Scientific®, Carlsbad, California, USA).
Beta-actin amplification was performed as an internal (endogenous) control of extraction, using Actin-FWD (5′-CGGAACCGCT-CATTGCC-3′) and Actin-REV (5′-ACCCA-CACTGTGCCCATCTA-3′), according to Ferreira et al. (2010). All blood samples were positive for this gene.
We found Leptospira spp. antibodies in two species of armadillo with a prevalence of 31% (5/16; 95% confidence interval [CI]=0.11–0.58) in yellow armadillo and 18% (4/22; 95% CI=0.05–0.40) in giant armadillo; titers ranged from 200 to 1,600 (Table 2). The specimens of nine-banded armadillo and southern naked-tailed armadillo tested did not present Leptospira spp. antibodies. Cross-reactivity occurred with serogroups Pomona and Icterohaemorrhagiae in sera of one giant armadillo, and we considered the highest titer as positive. All samples were negative by RBT. Blood samples from all individuals were PCR-negative for Leptospira spp. and Brucella abortus.
This study found that the Leptospira spp. seroprevalence in yellow armadillos and giant armadillos was moderate to low (31% and 18%, respectively) and related to three serogroups: Autumnalis, Cynopteri, and Pomona. The Pomona serogroup was the most frequently identified in our study. The serogroups found in our investigation were different from those most commonly detected in cattle herds in Pantanal. Many epidemiologic studies have detected the serogroup Sejroe serovar Hardjo as the most frequent in cattle from Pantanal (Miashiro et al. 2018).
Research suggests that titers between 100 and 200 characterize a low positive result and indicate exposure to leptospires, while titers equal or above 400 indicate recent exposure or active infection (Boqvist et al. 2002). In this study, a giant armadillo had a titer of 1,600 for serogroup Pomona, while two other individuals presented titers of 800 for serogroups Pomona and Cynopteri; by PCR, we did not find the agent in the bloodstream, and, at the time of capture, the individuals had no symptoms of leptospirosis, so we feel that the high titers indicated recent contact with the pathogen.
The serologic and molecular results obtained in our study indicate that the armadillos we evaluated did not have previous contact with Brucella spp. However, only blood samples were tested by PCR. Furthermore, the influence of massive brucellosis vaccination in the bovine herd present in the study area may have had a role, as demonstrated by Chate et al. (2009) and Filho et al. (2016), which showed that 37% of the herds were infected in 1998, but this decreased to 25% in 2009 due to the vaccination program implemented in 2004.
We did not find pathogenic leptospiral DNA by conventional PCR in the bloodstream of tested armadillos. However, our findings demonstrate that giant armadillos and yellow armadillos had contact with Leptospira spp. in the study area, although we found different serogroups from those commonly found in cattle in Pantanal, despite close contact between wildlife and cattle. Although suggestive, our results have many limitations, and future research is required to clarify the role of armadillos in the epidemiologic chain of leptospirosis and its possible impact in armadillo populations.
This study is part of the Giant Armadillo Conservation Program, which benefited from multiple grants, mostly from North American and European zoos, through the Institute of Wildlife Conservation in Brazil.