Capybaras (Hydrochoerus hydrochaeris) are the world's largest rodents and play an epidemiologic role in the transmission of zoonotic pathogens, including the causative agents of Brazilian spotted fever, leptospirosis, and others. We surveyed the health of 31 free-ranging capybaras at the Alberto Löfgren State Park, São Paulo, Brazil using a variety of diagnostic methods. Hematology and serum chemistry were consistent with mild malnutrition and parasitism but did not indicate severe physiologic imbalance or disease. All animals were serologically negative for Rickettsia rickettsii, Leishmania spp., and Trypanosoma sp., but antibodies against rabies virus (71%), Leptospira sp. (26%), and Toxoplasma sp. (23%) were detected. Salmonella sp. was not cultured from fecal samples. Frequently cultured enterobacteria included Escherichia coli (61%), Enterococcus casseiflavus (35%), Enterococcus faecalis (35%), Enterobacter aerogenes (32%), Klebisella pneumoniae (32%), and Serratia marcescens (32%). No potentially pathogenic fungi were cultured from hair samples. Fecal parasitology revealed infection by Protozoophaga sp. (58%), Viannella spp. (23%), Strongyloides spp. (10%), and Ancilostomatidae (10%). A total of 218 ticks was retrieved from the animals: Amblyomma sp. larvae and nymphs (43%), A. dubitatum adults (52%), and A. cajennense adults (5%). The capybaras were free from most potentially zoonotic pathogens evaluated; however, the presence of Amblyomma spp. ticks (potential vectors of Rickettsia spp.) and indirect evidence of exposure to the rabies virus, Leptospira sp., and Toxoplasma sp. warrant the maintenance of public health programs and wildlife health monitoring.
The capybara (Hydrochoerus hydrochaeris), the world's largest rodent, is broadly distributed in South America (Paula et al. 2002). Their high reproductive capacity, generalist feeding habits, and minimal habitat quality requirements, coupled with widespread population decline of its natural predators, have contributed to overpopulation of capybara in numerous regions (Pereira and Eston 2007). Therefore, monitoring these animals in areas cohabited by humans is important for adequate population management, with the aim of reducing potentially negative ecologic impacts on both species. In most wildlife populations, predation and malnutrition are the leading causes of mortality, whereas infectious diseases play a secondary role (Gonzáles Jiménez 1995).
Capybaras are particularly well known for their epidemiologic role in the transmission of Rickettsia rickettsii through the maintenance of Amblyomma spp. of ticks (MOH 2008). In addition to rickettsia, capybaras may also carry other potentially zoonotic agents, such as Leishmania spp., Leptospira spp., Trypanosoma spp., Enterobacteriae (including Salmonella spp.), ectoparasites, dermatophytes, and rabies virus (Muzel unpubl.). Capybaras may be an important proxy to monitor circulation of Toxoplasma gondii in wildlife communities and act as intermediate hosts in the parasite life cycle (Truppel 2009).
Along with indirect (serology) and direct diagnostic methods (bacteriology, mycology, and parasitology tests), hematology and serum chemistry provide auxiliary and flexible strategies to detect infectious diseases and to provide information on general health status, nutrition, etc. (Madella et al. 2006). We surveyed the health of free-ranging capybaras using several diagnostic methods, focusing on the most common zoonoses for these animals.
MATERIALS AND METHODS
In May 2011 we sampled 31 free-ranging capybaras at the Alberto Löfgren State Park, São Paulo, Brazil. These represented all capybaras in the park, apart from one adult male of a single social group: seven adult males, nine adult females, five subadult males, nine subadult females, and two female cubs. Animals were chemically restrained using ketamine and xylazine (5 mg/kg and 0.1 mg/kg, respectively, on the basis of visual estimates of body mass) administered intramuscularly via blow dart. After capture, animals were weighed and ketamine and midazolam (3 mg/kg and 0.5 mg/kg, respectively) were administered intramuscularly with a syringe. Venipuncture of the femoral vein was used to collect 10 mL of blood per individual; blood volume was separated in tubes with ethylenediaminetetra-acetic acid and without anticoagulants. Fecal samples were freshly collected from the rectum and separated into tubes with 2.5% potassium dichromate and sterile tubes. Hair samples from the dorsum were collected with sterile instruments and stored in sterile tubes. Blood, fecal, and hair samples were transported by ice packs and processed within 2–6 hr. A standardized 5-min search for ticks was conducted on one side of the body and all ticks found were collected using hemostats.
Hematology and serum chemistry
Routine laboratory methods were used to determine the following hematologic parameters: hematocrit (Hct, %), hemoglobin (g/dL), red blood cell count (RBC, 106 cells/µL), mean corpuscular volume (MCV, fL), mean corpuscular hemoglobin (pg), mean corpuscular hemoglobin concentration (g/dL), white blood cell count (WBC, cells/µL), heterophils (cells/µL), lymphocytes (cells/µL), monocytes (cells/µL), eosinophils (cells/µL), and basophils (cells/µL). Erythrocytes and leukocytes were counted manually in a Neubauer chamber, after dilution of the blood sample in Gower's (1∶200) and Türk's (1∶20) solutions. The differential leukocyte counts were performed in fresh blood smears that were stained with Diff-Quick stain, by counting 100 cells on an optical microscope. Hemoglobin concentration was measured using the cyanmethemoglobin method (Drabkin and Austin 1935).
Serum chemistry parameters, including total plasma protein (PT, g/dL), albumin (ALB, g/dL), blood urea nitrogen (BUN, mg/dL), alanine transaminase (ALT, U/L), gamma-glutamyltransferase (GGT, U/L), and alkaline phosphatase (ALP, U/L) were obtained in a semiautomated system Drake (Quick Lab®; Biotécnica Ltda, Varginha, Minas Gerais, Brazil), using commercial kits (Labtest®; Lagoa Santa, Minas Gerais, Brazil). Creatinine (mg/dL) was obtained by the Jaffé method, using a commercial creatinine assay kit (DiaSys®; Diagnostic Systems, Holzheim, Germany). Globulin concentration (g/dL) was calculated as the difference between PT and ALB concentrations.
Indirect immunofluorescence was used to test serum samples for Leishmania sp., R. rickettsii, Toxoplasma sp., and Trypanosoma sp. (D'Auria et al. 2010). Serum neutralization was used to test for the rabies virus (Smith et al. 1973). Microscopic agglutination was used to test for the following Leptospira spp. serovars (WHO 2003): andamana, australis, autumnalis, batavie, brasiliensis, butembo, canicola, castellonis, copenhageni, cynopteri, djasiman, grippotyphosa, hardjo, hebdomadis, icterohaemorrhagiae, javanica, panama, patoc, pomona, pyrogenes, serjoe, hermani, tarassovi, and wolffi. The cutoff point was 0.5 IU/mL for rabies virus (WHO 1994), 1∶16 for Toxoplasma sp. (Camargo 1964), and 1∶100 for Leptospira sp. (WHO 2003).
Fecal samples for the culture of Salmonella sp. were transferred to tubes containing 0.1% peptone water and incubated at 35 C for 24 hr. After incubation, 0.1 mL of each sample was transferred to an assay tube containing 10 mL of tetrathionate broth and incubated at 37 C for 24 hr. Next, 0.1 mL of the bacterial culture was seeded on xylose lysine deoxycholate agar and incubated at 37 C for 24–48 hr. Bacterial identification was based on colony morphology and biochemistry. Species identification was carried out using conventional biochemical tests (Newprov®, Pinhais, Paraná, Brazil), EPM (glucose fermentation, production of gas and H2S, urea hydrolysis, and tryptophan deaminase reaction), MILi (mobility, indole, and lysine decarboxylase), and Simmons citrate (Holt et al. 1994).
Fecal samples for the culture of Enterobacteriaceae were seeded in brain–heart infusion broth and incubated at 35 C for 24 hr. Cultures were transferred to blood–MacConkey–sorbitol agar plates and incubated at 37 C for another 24–48 hr. Bacterial identification was based on colony morphology, biochemistry (Holt et al. 1994), and automated turbidity analysis (Silva and Neufeld 2006). Colonies identified as Escherichia coli were subjected to further analysis: DNA was extracted (Boom et al. 1990) and a multiplex PCR was used to detect the genes eae (917 base pairs [bp]) and bfp (326 bp) (Aranda et al. 2007). Postamplification fragments were visualized using 1.5% agarose gel electrophoresis with Blue-Green stain (LGC Biotecnologia, São Paulo, Brazil).
Hair samples were placed in Sabouraud broth and incubated at 35 C for at least 15 days. Samples were then seeded in chloramphenicol–Sabouraud–dextrose and Micosel agar plates and incubated at room temperature and at 37 C. Growth of fungal colonies was verified every other day and identification was based on macroscopic and microscopic morphology (Muzel, unpubl.).
Fecal samples in potassium dichromate were evaluated through classic fecal parasitology methods including Willis and Molay, Sheather's sugar centrifugal flotation, and Ritchie technique (Cimermam and Franco 1999). Ectoparasites were identified using the dichotomic and pictorial keys of Linardi and Guimarães (2000).
With the exception of RBC and WBC (n = 29) and serum chemistry (n = 30), sample size was 31 for all analyses. General linear models were used to determine whether sex (male, female) and age group (cub, subadult, adult) influenced hematologic parameters. General linear models were also used to determine whether these blood parameters correlated with serology (rabies, Toxoplasma sp., Leptospira sp.) and fecal parasitology. Post hoc Tukey comparisons were used to identify differences between categories. Fisher's exact tests were used to determine whether positive serology (rabies, Toxoplasma sp., or Leptospira sp.) or fecal parasitology results were unevenly distributed among sexes or condensed age groups (cub+subadult, adult). The significance level was 0.05 for all tests.
Average body mass was 56.5±12.66 kg (mean±SD) for adult males, 57.2±6.30 kg for adult females, 38.6±7.41 kg for subadult males, 34.8±14.98 kg for subadult females, and 28.2±1.76 kg for female cubs. Hematologic and serum chemistry parameters (Table 1) did not differ significantly in relation to sex or age group (all P>0.1), with the exception of lymphocytes (F = 10.72, P<0.001) and basophils (F = 12.12, P<0.001) in relation to age group, and urea nitrogen (F = 7.4, P = 0.011) in relation to sex. Post hoc comparisons indicated that cubs had higher counts of lymphocytes (mean±SD = 6,635±3,302 cells/µL) and basophils (353±109) than did subadults (2,526±1029 and 18±60) or adults (2,078±1,074 and 61±93), and that males had higher BUN (45.3±10.5 g/dL) than females (37.3±5.7).
Several hematologic values (Hct, Hb, RBC, and MCV) and serum chemistry (albumin, total protein, creatinine, ALT, GGT, and ALP) were lower than reference parameters published by the International Species Information System (ISIS 2002).
All animals were serologically negative for R. rickettsii, Leishmania sp., and Trypanosoma sp., whereas positive serology was observed for rabies virus (22/31, 71%) and Leptospira sp. (8/31, 26%). All animals were negative for Toxoplasma sp. using indirect hemagglutination; however, 23% (5/31) of animals were positive using indirect immunofluorescence. Positive results for rabies virus, Leptospira sp., and Toxoplasma sp. serology were evenly distributed among young animals and adults (all P>0.05). For rabies virus, individuals had antibody concentrations equal to 0.5 IU/mL, 17 had concentrations between 0.66 and 1.3 IU/mL, and three individuals had concentrations higher than 1.3 IU/mL. Table 2 provides details of Leptospira sp. serotypes. Tables 3 and 4 summarize the results for fecal bacteria culture and parasitology. Escherichia coli was cultured from 19 fecal samples (61%), but no strains were positive for the pathogenicity factors eae and bfp. Hair fungal culture revealed the growth of nonpathogenic mycelian fungi such as Penicillium sp. (30% of individuals), Cladosporidium sp. (25%), Acremonium sp. (22%), Scopulariopsis sp. (6%), and Chrysosporium sp. (6%), and of yeast such as Candida spp. (20%) and Rodothorula sp. (4%). A total of 218 ticks was retrieved from the animals: 114 Amblyomma dubitatum (70 adult males, 44 adult females), 10 A. cajennense (five adult males, five adult females), and 94 Amblyomma sp. (four larvae, 90 nymphs).
No association was found between hematology and serum chemistry parameters and positive serology or fecal parasitology results, with the exception of individuals with positive serology for Leptospira sp., which had significantly lower RBC (F = 10.24, P = 0.004; positive: 2.42±0.04, negative: 2.91±0.27), Hct (F = 9.25, P = 0.006; positive: 33.38±5.53, negative: 37.71±2.19), and Hb (F = 9.49, P = 0.005; positive: 10.40±1.92, negative: 11.98±0.95), and individuals with positive results in fecal parasitology, which had lower MCV (F = 6.24, P = 0.020; positive: 129.29±9.80, negative: 137.94±9.14).
Hematology and serum chemistry provide parameters of the blood physiologic and immunologic responses, which indirectly reflect an animal's health status and may suggest the presence of infectious and non-infectious diseases (Muñoz and Montoya 2001). However, studies on hematologic parameters and serum chemistry in capybaras are scarce and often result from small sample sizes, potentially skewing the interpretation. Our results for several hematologic values (Hct, Hb, RBC, and MCV) were lower than those of other published studies in capybaras, which may suggest the presence of microcytic anemia, probably associated with iron deficiency (ISIS 2002). By dealing with free-ranging animals, it is difficult to justify the theoretical cause to blood loss, but one hypothesis, also supported by the rabies serology results, is that these animals participate in the life cycle of vampire bats, which are present in the studied region.
Furthermore, even though total leukocyte counts were not elevated, relative and absolute eosinophilia were observed. Eosinophils play a role in the immune response to tick parasitism in capybaras (Heijden et al. 2003). Additionally, basophil counts increase due to the same causes as eosinophils (parasite infestation or fungal infection, neoplasia, hypersensitivity reaction). Thus, these animals may have increased basophils, along with eosinophils, due to parasites, which were present in most of the examined animals (Table 1). Lymphocyte and basophil counts were higher in young animals of this study, when compared with reference values (ISIS 2002). Lymphocyte numbers vary with age or species. Absolute counts are higher in juveniles of domestic species (e.g., cats; Felis catus); transient lymphocytosis sometimes occurs with excitement or exercise (especially in horses; Equus ferus caballus) (Harvey 2012).
In comparison with published studies (e.g., Reyes et al. 2009; Corredor-Matus and Rodriguez-Pulido 2010), serum chemistry also revealed mildly decreased values of albumin, total plasma protein, creatinine, ALT, GGT, and ALP. None of the studied animals had findings indicative of severe physiologic imbalances or clinical disease associated with typical causes such as acute inflammation, hepatic insufficiency, renal disease, or severe gastrointestinal or skin diseases. However, both juveniles and adults had mild malnutrition and parasitism that could be associated with hematologic and serum chemistry imbalance parameters.
We did not observe serologic evidence of exposure to R. rickettsii, Leishmania sp., or Trypanosoma sp. Rickettsia rickettsii is the causative agent of the Brazilian spotted fever, an emerging zoonotic disease causing high human fatality that has been recorded in southern, southeastern, and northeastern Brazil, including the state of São Paulo (MOH 2011). Even though our capybaras had not been exposed to this pathogen, capybaras are known to play an important role in its epidemiology through the maintenance of ticks, particularly A. cajennense (Pereira and Eston 2007). In fact, large numbers of ticks were observed, including A. cajennense, indicating that the potential for the transmission of R. rickettsii exists in the studied population and monitoring should continue. Despite the epidemiologic role played by other rodents in the transmission of leishmaniasis (Ready et al. 1983), neither this nor previous studies (Valadas et al. 2010) have found evidence of exposure of capybaras to Leishmania sp. However, Trypanosoma spp. may be found in a broad variety of wild mammals, including capybaras (Mazzei et al. 2009). For example, T. evansi was isolated in 24% of free-ranging capybaras in Colombia (Morales et al. 1976), and antibodies against T. cruzi were present in 8% of free-ranging capybaras in the state of São Paulo (Valadas et al. 2010). Although none of the capybaras we examined had antibodies against Trypanosoma sp., preliminary investigations have detected antibody to T. cruzi in opossums (Didelphis sp.) at Alberto Löfgren State Park (Mazzei et al. 2009).
There are no standard cutoff points for the interpretation of rabies virus serologic tests in capybaras; therefore, we used the cutoff point applied for humans (WHO 1994). Using this criterion, 71% of the studied capybaras were antibody positive. As preliminary investigations have also detected antibody in nonhematophagous bats at Alberto Löfgren State Park (Mazzei et al. 2009), our results emphasize that the area may still present a risk of the exposure to humans. This is particularly true considering the proximity (<2 km) of the state park to densely populated urban areas.
Twenty-three percent of the capybaras were positive for Toxoplasma sp. using the indirect immunofluorescence test. Only felids are considered definitive hosts, because the sexual phase cycle of Toxoplasma sp. is dependent on development in the intestines of these animals. Nonfelids, including capybaras, act as intermediate hosts and may contribute to the circulation of this pathogen if they or their carcasses are consumed by felids. Capybara meat (obtained via hunting) containing cysts may pose risks to human health. Additionally, the foraging habits of capybaras and their adaptation to human-affected areas make them valuable proxies for monitoring circulation of Toxoplasma in wildlife communities (Yai 2007).
As rodents, capybaras play a role in the maintenance and transmission of Leptospira sp. (Pimentel et al. 2009). The interpretation of indirect serologic tests for Leptospira sp. may be challenging, as the diagnosis of active infections generally requires collection of consecutive samples with the observation of fourfold or more increase in titer. Furthermore, to determine which Leptospira sp. serovar is involved, the criteria state that the specific serovar should have a serologic titer ≥1∶800 or, ideally, 1∶1,600 (Nogueira and Cruz 2007). Previous studies in Venezuela and Brazil (Paula, unpubl.) identified the serovars canicola, copenhageni, balllum, hardjo, hendomadis, icterohaemorrhagiae, and wolffi as the most frequent. In contrast, we only found high titers (>1∶800) for serovars grippotyphosa, hardjo, and pomona.
Salmonella sp. was not identified in the feces of the studied capybaras. Enterobacteriae that are known to be opportunistic human pathogens, such as Aeromonas sp., Citrobacter freundi, Escherichia coli, and Proteus sp., were retrieved (Sarkis 2002). Escherichia coli, in particular, may be pathogenic when it acquires virulence factors (i.e., genes that confer an increased ability to colonize and produce infection or evade the host's defense mechanisms; Trabulsi and Alterthum 2005). Although we cultured E. coli from 61% of capibaras, we did not find evidence of enteropathogenic E. coli virulence factors (eae and bfp) in any of the strains. No dermatophytes or other potentially pathogenic fungal species were cultured from the hair of the capybaras, and the few yeast organisms obtained (Candida spp. and Rodothorula sp.) were likely to be contaminants rather than colonizers.
More than 80 endoparasitic species have been identified in capybaras (Salas and Herrera 2004), which may be largely related to the gregarious, amphibious, and territorial habits of the species (Sinkoc et al. 1997). The parasites observed in this study were Protozoophaga sp. (58%), Viannella spp. (23%), Strongyloides spp. (10%), and Ancilostomatidae (10%); Costa and Catto (1994) observed the same species and patterns, but with higher frequencies for most species.
Although the capybaras we studied were free from most potentially zoonotic pathogens commonly associated with the species, the presence of Amblyomma ticks (potential vectors of Rickettsia sp.) and indirect evidence of exposure to the rabies virus, Leptospira sp., and Toxoplasma sp. warrant a public program to manage and discourage contact between these animals and nearby human populations.
We are thankful to Ana Lucia Arromba and to the staff of the Alberto Löfgren State Park, to the Laboratory of Wildlife Comparative Pathology and to the Laboratory of Clinical Analyses (Universidade Paulista).