Study areas
This study was performed in two forest fragments in the Brazilian Midwest: (i) a conservation unit called Prosa State Park (20°26’59’’S, 54°33’55’’O), and (ii) a residential area known as Air Force Village (20°28’17’’S, 54°39’14’’O) (Fig. 1).
Capture of coatis and sample collection
Animals were captured and recaptured from March 2018 to May 2019 using box traps (90×45×50; Equipos Fauna®, Brazil), sedated with 7 mg/kg of Telazol anesthetic (Zoetis®, USA), and marked with subcutaneous microchips (Animal Tag®, Brazil). After sedation, animals were clinically inspected, and their dental conditions and body size measurements were recorded to estimate age through model-based analysis41. Blood was aseptically collected from the femoral vein using 0.25 × 8 mm needles (BD® Vacutainer) and deposited in tubes containing ethylenediamine tetraacetic acid (EDTA), as well as tubes containing clot activator to obtain serum. Bone marrow was aseptically obtained from the manubrium sterni using 0.40 × 1.2 mm hypodermic needles and 10 mL syringes and placed in EDTA tubes. All biological material was kept at 4°C in thermal boxes until laboratory procedures. After complete sedation recovery, animals were released at capture sites. All field procedures were conducted in accordance with the Instituto Chico Mendes de Conservação da Biodiversidade (license number 56912-2), Instituto de Meio Ambiente de Mato Grosso do Sul (IMASUL license number 05/2017, process Nº61/405959/2016), Air Force Cooperation Agreement (Nº01/GAP-CG/2018), and Ethics Committee for Animal Use of Universidade Católica Dom Bosco, CG, MS (license number 001/2017).
Hematological and biochemical analysis
Red blood cell (RBC), white blood cell (WBC), hemoglobin (Hgb), hematocrit (Hct), mean corpuscular volume (MCV), mean corpuscular hemoglobin concentration (MCHC) and platelet counts were measured by an automated hematology analyzer (POCH-iV 100, Syxmex®, Brazil) according to manufacturer’s recommendations. Validation of the Diff pocH-100iV hematological analyzer for use in coatis was performed by a specialized Sysmex technician using blood from five clinically healthy adult coatis together with hematological values for N. nasua reported in the International Species Identification System42. Standard quality controls consisted of low, normal, and high levels of animal platelet, RBC, and WBC material (EIGHTCHECK-3WP X-TRA, Sysmex®), performed daily before each analysis. Different leukocyte counts were performed in blood smears fixed with methanol, stained with Panótico kit (Laborclin®, Brazil), and visualized using a binocular microscope (Eclipse E200, Nikon®, Japan). Serum concentrations of creatine kinase (CK), CKMB fraction, lactate dehydrogenase (LDH), alkaline phosphatase (AP), alanine aminotransferase (ALT), aspartate transaminase (AST), urea, creatinine, albumin, total serum protein (TSP), total cholesterol, high-density lipoprotein (HDL) cholesterol, glucose, and triglycerides were determined by spectrophotometer analysis (Brasmed®, Brazil) using the Biotech® and Gold® commercial kits. Globulin levels were determined based on the difference between TSP and albumin levels.
L. infantum detection
Diagnosis of L. infantum infection was based on parasitological, serological, and molecular methods. Bone marrow samples were placed in biphasic media, Novy-MacNeal-Nicolle (NNN), with Schneider’s insect medium overlay supplemented with 10% fetal bovine serum. Tubes were incubated at 26°C–28°C and examined weekly for up to 30 days. Detection of anti-L. infantum IgG and K26/K39 specific antibodies was performed through the enzyme-linked immunosorbent assay (ELISA modified from EIE®, BioManguinhos, Brazil) and the dual path platform chromatographic immunoassay (DPP® CVL Biomanguinhos, Brazil), respectively, as described by de Macedo et al.8. Molecular detection of L. infantum was conducted using DNA extracted from bone marrow samples, following the protocol described by de Macedo et al.8. DNA samples were first tested by PCR for Leishmania spp. kinetoplastid DNA43. Positive samples were confirmed by PCRs for heat-shock protein 70 (hsp70) and L. infantum kDNA genes44,45. The hsp70 PCR products were sequenced using the Sanger method on an AB3500 platform (Applied Byosystems®, United States), and the obtained sequences were deposited in GenBank database (accession numbers are described in de Macedo et al.8.
Data analysis
To avoid misinterpretation, only adult individuals were included in the analysis, and capture and recapture (at least one month apart) were considered independent. Descriptive statistics (mean ± standard deviation) were used to obtain mean hematological and serum biochemistry values of the uninfected (UN) and infected (LI) animals. A coati was considered infected when its samples were positive in at least one diagnostic method (parasitological, molecular, and serological, if positive in both ELISA and DPP®CVL), as described in de Macedo et al.8. The Shapiro–Wilk test was used to check the normality of the hematological and biochemical parameters, and the Student’s t-test and Mann–Whitney U-test were performed according to data distribution to observe differences between the UN and LI groups.
To determine the indirect impact of L. infantum infection on BC through health indicators, we performed path analysis according to Santos et al.18. We estimated BC based on residuals from an ordinary linear regression between body weight and head-to-body length of the coatis. Health indicators were categorized into: (i) red blood cell (RBCI) (RBC, Hgb, Hct, MCV, and MCHC); (ii) coagulation (COI) (platelets); (iii) immune response (IMRI) (globulin and lymphocytes); (iv) infection response (IRI) (monocytes, neutrophils and eosinophils); (v) kidney damage (KDI) (urea and creatinine); (vi) liver damage (LDI) (AST, ALT, and AP); (vii) cardiac damage (CDI) (CKMB); (viii) skeletal muscle damage (SMDI) (CK and LDH); (ix) nutritional profile (NPI) (triglycerides, total cholesterol, HDL cholesterol, and glucose); and (x) protein profile (PPI) (albumin and TSP). Variables were considered statistically significant for p values < 0.05. All data were analyzed using R software46.