Detection of zoonotic enteropathogens in captive large felids in Italy

Within the One Health paradigm, infectious disease surveillance have been developed for domestic and wild animals, leaving the role of captive non‐domestic populations, especially felids in zoos and circuses, less explored. This study addresses the proximity of these captive animals to urban areas, necessitating focused monitoring for potential zoonotic enteropathogens. The present work aimed to investigate the presence of such zoonotic enteropathogens in faecal samples from captive large felid populations.


| INTRODUC TI ON
Captive felid populations are exposed to a variety of potentially pathogenic microorganisms and parasites (Terio et al., 2018).
Sources may be feeding with contaminated raw meat, hunting some small prey species with free access to the boxes and cages, such as small wild rodents and birds, presence of cats in the zoo that can spread pathogens by their faeces in zoo enclosures and food contamination due to improper storage.While the reclusive nature of captive animals may limit the transmission of infectious diseases, there may not be individual or population immunity making them vulnerable to severe and lethal infections (Terio et al., 2018).
In turn, captive felids can shed a wide range of potential pathogens in faeces and other bodily secretions that can persist in the environment for extended periods (Hatam-Nahavandi et al., 2021;He et al., 2018;Javaregowda, 2016).This is exacerbated in zoos and circuses because of the high concentration of animals in the same enclosures, poor hygiene and stress associated with captive conditions.
Large Felidae are kept in captivity in zoos or circuses in proximity to urban areas and thus in close contact with people, particularly animal keepers, veterinarians and visitors.They may act as a source of zoonotic infections by direct contact, faecal-oral transmission, indirect contact with insect vectors and contaminated inanimate objects or inhalation of infectious materials (Chomel et al., 2007).
The Centers for Disease Control and Prevention have provided information on diseases that can be passed from wild carnivorous species to people, including rabies, ringworm and external para- Another serious concern is that captive felid populations can harbour antimicrobial-resistant pathogens as a result of the close anthropogenic pressures, contact with domestic animals and meat contaminated by drug-resistant organisms (Ghosh et al., 2019).For this reason, the role of captivity in the maintenance and dissemination of antimicrobial resistance has received increased attention in recent literature (Baros Jorquera et al., 2021).
The development of appropriate programmes for monitoring zoonotic pathogens in animal reservoirs is essential.However, while surveillance programmes exist for farm animals, few or no programmes are specifically aimed at wildlife and exotic pets in captivity.
In this context, we investigated the presence of known infectious agents of zoonotic diseases, specifically Salmonella spp.and multidrug-resistant strains and gastrointestinal parasites, in captive Felidae populations housed in zoos, circuses and rescue centres in Italy.

| Study sites and animal sampling
One hundred and eight faecal samples of captive large felids were collected in three circuses (C1, C2 and C3), five zoos (Z1, Z2, Z3, Z4 and Z5) and a rescue centre (RC) across Italy between 2020 and 2021.
None of the specimens was obtained invasively.Faecal material was collected from the ground of the enclosures and placed in sterile collection containers.All animals were asymptomatic/subclinical at the time of sampling.However, in general anamnesis, animal keepers of all collection sites reported sporadic and self-limiting episodes of diarrhoea in the last 6 months but none was being managed clinically for diarrhoeic syndrome.Ethics approval was not necessary for this study as the faecal samples were collected non-invasively for routine clinical examinations.
Since eight animals were referred as 'black panthers', which is not a taxonomic entity but rather a common name to identify the melanistic forms of leopards (Panthera pardus, Linnaeus, 1758) and jaguars (Panthera onca Linnaeus, 1758), a double enzyme-based PCR-RFLP test was designed to provide reliable taxonomic identification of the samples.Molecular identification was not necessary for all other species, as morphological features and coat coloration were sufficient to produce a specific attribution.DNA isolation from faeces of all the ambiguous samples was obtained by using the QIAamp® Fast DNA Stool Mini kit (QIAgen, Milan, Italy) following the manufacturer's instructions.Leopard and jaguar sequences of the mitochondrial gene cytochrome oxidase 1 (COI) were amplified by using 12.5 μL of 2× PCR MasterMix (Promega, Madison, WI, USA), 1 μL of each of the multi-taxa primers 5′-NEST-F and 3′-NEST-R (F 5′-GTAAT CGT TAC AGC CCATGC-3′; R 5′-GGGTC GAA AAA TGT GGTGTT-3′) (Gandolfi et al., 2017), 2 μL of the priorly extracted DNA and NucleaseFree

Impacts
• Asymptomatic carriers of zoonotic enteropathogens appear to be common within captive felid populations and likely pose an infection risk to occupationally exposed people and the general public.
• Salmonella Infantis, Enteritidis, Typhimurium, Derby, Newport and Coeln, mainly responsible for human salmonellosis, were all isolated from faecal samples of captive felids.
• A high number of Salmonella enterica subsp.enterica serovariants resistant to both ampicillin and trimethoprimsulfamethoxazole were observed.
• A scant presence of Giardia duodenalis and soil-transmitted helminths was detected; the exposure of large felids to Toxoplasma gondii should be better investigated through serological surveys.
H 2 O (Promega, Madison, WI, USA) to a final volume of 25 μL.DNA restriction was performed at 37°C for 3 h on a total volume of 20 μL as follows: 10 μL of the amplified product; 2 μL of restriction enzyme HhaI (New England Biolabs) and BglII (Promega, Madison, WI, USA) were chosen to digest leopard DNA and jaguar DNA respectively); 2 μL of enzymatic buffer; and 6 μL of NucleaseFree H 2 O (Promega, Madison, WI, USA).Restriction products were checked on a 3.5% agarose gel electrophoresis run in 1× TBE buffer.
In all sites, animals were fed with raw poultry meat, sporadically replaced with rabbit or beef.Faecal samples were then refrigerated at 4°C and shipped within 24 h to the Veterinary Microbiology and Parasitology Laboratories of the University of Perugia.

| Bacteriological investigations and antibiotic susceptibility testing
One gram of each faecal sample was placed into a culture tube containing 9 mL of buffered peptone water (Liofilchem, Roseto degli Abruzzi, Italy) and incubated at 37°C for 20 h.One hundred μL of the pre-enriched culture was transferred in 10 mL of Rappaport

| Parasitological investigations
Sixty out of the 108 faecal samples collected were also examined for the presence of gastrointestinal parasites.For this purpose, each sample was firstly examined macroscopically, to verify the presence of nematodes, cestodes and/or fragments of parasites, and then was divided into two aliquots (approximately 10 g each).The first aliquot was tested by 33% zinc sulphate centrifuge-flotation technique (specific density 1180), and the slides were checked under the microscope at 100×, 400× and 1000× magnifications for the detection of nematode and protozoan cysts, oocysts and nematode eggs, identified by standard parasitological criteria.The second aliquot was suspended in 10 mL of sterile phosphate-buffered saline, filtered using a stainless-steel mesh sieve (pore size 100 μm) and centrifuged at 1500 g for 10 min in polypropylene tubes.The supernatant was discarded and the sediment was resuspended in 500 μL of PBS and two aliquots of 20 μL and 200 μL of the suspension were used, respectively, for direct immunofluorescent staining assay (DFA), using a commercial kit (MeriFluor Crypto&Giardia; Meridian® Bioscience Inc., Cincinnati, OH, USA) and for genomic DNA extraction using the QIAamp® Fast DNA Stool Mini kit (QIAgen, Milan, Italy), according to the manufacturer's instructions with five additional min freeze/ thaw cycles before the DNA extraction to destroy the cysts/oocysts.PCR assays targeting a 292-bp fragment of the small subunit rRNA gene (SSU-rRNA) of Giardia and a 130-bp fragment of the B1 gene of Toxoplasma gondii were performed as described previously (Hopkins et al., 1997;Lin et al., 2000).The amplified product obtained from the PCR for Giardia was directly sequenced in both directions using a 16-capillary ABI PRISM 3130 × l Genetic Analyzer, assembled and edited with SeqScape software v 2.5 (Applied Biosystem, Foster City, CA, USA).The assembled sequences were compared with homologous SSU-rRNA sequences for Giardia available in GenBank, using the Basic Local Alignment Search Tool (http:// www.ncbi.nlm.nih.gov/ BLAST/ ) and then aligned with representative sequences using MegaX software.

| Statistical analysis
After running descriptive statistics for frequency observations with a 95% confidence interval, inferential statistical analysis was performed to compare the prevalence of Salmonella isolates and gastrointestinal parasites in different study sites, as appropriate.

| Parasitological results
The parasitological results associated with the animal hosts and housing sites are shown in Table 3.Five out of the nine sites of sampling (RC, C2, C3, Z1 and Z5) tested negative for parasitic infections.
Overall, only parasitic infections by Toxascaris leonina and Giardia duodenalis were detected.
Toxascaris leonina eggs were microscopically observed in 11 faecal samples (18.33%, 95% CI 9.5%-30.5%)out of the 60 analysed; Giardia cysts were detected by DFA in one felid (P.pardus) (1.66%, 95% CI 1%-8.9%)housed in Z2.The DNA template of the specimen that tested positive from the DFA was successfully amplified by Giardia-specific PCR protocol.The amplicon of the expected size was repeatedly obtained and the BLAST analysis of the sequence showed an identity score from 99%-100% with available sequences TA B L E 1 Salmonella enterica subsp.enterica serovariants isolated from faecal samples of captive wild felids housed in the three circuses (C1, C2 and C3), four zoos (Z1, Z2, Z3 and Z4) and a rescue centre (RC) in Italy.

| DISCUSS ION
For almost 20 years, the One Health initiative has constituted a global strategy to control the circulation of zoonotic infectious diseases by acting on the human-animal-ecosystem interface.In this context, surveillance programmes have been developed for domestic animals and more recently wild fauna, less considering captive non-domestic populations.However, knowledge of the role of captive animals in the One Health era is growing.Some zoo animal collections were utilized as sentinels of diseases for people, wildlife and domestic animals in nationwide disease surveillance systems (Deem, 2015).
Captive felids are popular in zoos and circuses, where their proximity to urban areas demands for monitoring efforts on infectious diseases that have the potential to be transmitted to nearby people and domestic animals.Among these, enteropathogens are relevant due to their zoonotic potential and high environmental contamination.
Faecal shedding of Salmonella in exotic felids was previously observed (Clyde et al., 1997).However, relevant data about the prevalence, pathogenicity and distribution of Salmonella spp. in large felids are scarce.The most prevalent serovariant isolated in this work was S. enterica subsp.enterica serovariant Infantis, followed by Coeln and Newport.Some previous results reported S. enterica subsp.enterica serovariant Typhimurium as the most prevalent type isolated from faecal samples of captive carnivores (Clyde et al., 1997;Venter et al., 2003), while we detected S. enterica subsp.enterica serovariant Typhimurium only in Z3.Overall, in literature there has been considerable disagreement with regard to the prevalence of Salmonella spp.
isolated from faeces of captive Felidae (Berardi et al., 2014;Lewis et al., 2002).This is likely related to the diet.The same Salmonella serovariant was indeed detected in the meat source and faeces of captive carnivores with little genomic variation (Venter et al., 2003).
The diet of captive wild animals commonly includes poultry products.Previously, chicken meat was identified as a source of strains of Salmonella isolated from people and captive wild animals (Smith et al., 2014) Centre for Disease Prevention and Control (ECDC), 2021).We thus hypothesize that poultry meat was the source of Salmonella for the captive animal collections examined, although it was not possible to collect meat samples to investigate microbial contamination.
Salmonella isolates can cause enteritis in non-domestic felids (Terio et al., 2018).However, here, the animals were asymptomatic/ subclinical at the time of sampling.This finding concurred with the concept that asymptomatic carriers appear to be common within captive populations and likely pose an infection risk for people (Hoelzer et al., 2011).In general anamnesis, animal keepers of all collection sites reported sporadic and self-limiting episodes of diarrhoea, anecdotally described as adverse feed reactions.On a wider level, research is needed to determine the virulence of Salmonella spp. in large felids.
Most of the isolated serovariants are pathogenic for people.
Particularly, the top five serovariants responsible for human salmo-   (Haeghebaert et al., 2001;Vestrheim et al., 2016).In light of the virulence features of the isolates, contacts with captive felids or environmental surfaces in public settings and exhibit environment, such as circuses and zoological parks, may represent a source of Salmonella infection for occupationally exposed people as well as the general public.Previous studies support the Salmonella risk posed to people, in fact zoonotic disease outbreaks were associated with captive animals in public settings and most of these were caused by Salmonella species (Bender et al., 2004;Steinmuller et al., 2006).The high number of isolates found resistant to one or more antibiotics is also of concern.This matches with the incidences of isolation of antibiotic-resistant Salmonella in poultry production (Castro-Vargas et al., 2020).However, no isolate was fully resistant to the fluoroquinolone enrofloxacin.This is comforting since fluoroquinolones are the drugs of choice to treat invasive salmonellosis in adult people.
Previous studies provided evidence that antimicrobial resistance determinants and pathogens are unwittingly common in captivity environments (Power et al., 2013).This may establish opportunities not only for the cross-animal species transmission of resistant pathogens but also for the transfer of resistance determinants be-  et al., 2011;Fagiolini et al., 2010).Our results confirmed this finding, although T. leonina was the only species of geohelminth identified here.Roundworms as Toxocara cati (González et al., 2007;Naydenko et al., 2012) or hookworms (e.g., Ancylostoma spp.), which are responsible for important human 'larva migrans' syndromes, were not recovered in the faecal samples; however, they were diagnosed in previous surveys conducted in Italian zoological gardens (Fagiolini et al., 2010).
No sample tested positive for T. gondii.Even if no epidemiologic data on the route of acquisition of T. gondii infection by captive felids in zoos or circuses are available, it is presumable to think that cyst-forming parasite is biologically adapted in these animals to be transmitted more efficiently by carnivorism than by faecal contamination, as observed for domestic cats.Raw chicken meat, which harbours T. gondii tissue cysts (Schares et al., 2018), is frequently offered to captive felids in the sampling sites; thus, the possibility of the animals to became infected should be considered.Also, the ingestion of infected rodents or birds that may have access within the animal enclosures or the mechanical transmission acted by flies, cockroaches or dung beetles entering the living area of the large felids may be implicated as additional routes of transmission (Thiangtum et al., 2006).The exposition of large felids to T. gondii should be better investigated through serological surveys (Iatta et al., 2020) since the oocysts shed by infected captive large felids may be a source of infection for further zoo animals, zoo keepers, zoo veterinarians and people visiting zoos (Thiangtum et al., 2006).Furthermore, it should be considered that acute clinical toxoplasmosis has been described sites, tularemia, brucellosis, leptospirosis, yersiniosis, campylobacteriosis, salmonellosis, cryptosporidiosis, giardiasis, infections with pathogenic Escherichia coli, toxoplasmosis, echinococcosis and cutaneous and visceral larval migrans (Centers for Disease Control and Prevention, 2021; Institutional Animal Care and Use Committee, Washington State University, 2021).
Vassiliadis (RV) Broth (Liofilchem, Roseto degli Abruzzi, Italy) and then incubated at 42°C for 24 h.Ten μL of RV broth culture was streaked on Chromogenic Salmonella Agar and Xylose Lysine Desoxycholate Agar (XLD agar) plates and incubated at 37°C for 20 h in aerobiosis.One colony per plate with morphology consistent with Salmonella was screened by urease test in Christensen Urea Agar and commercially available API identification systems (BioMérieux, Marcy-l'Étoile, France).Isolates identified as Salmonella were submitted for serotyping by agglutination test according to standards EN ISO 6579-3:2014.The antibiotic susceptibility profiles of 40 Salmonella were evaluated by determining the minimum inhibitory concentration (MIC).The selected isolates were considered to be representative of Salmonella serovariants circulating in each zoological structure.Broth microdilution was performed in triplicate according to CLSI standards VET01 to define the lowest antimicrobial concentration that inhibits visible bacterial growth (Clinical and Laboratory Standards Institute (CLSI), 2018a).According to the CLSI standards VET08 about faecal isolates of Salmonella spp.(Clinical and Laboratory Standards Institute (CLSI), 2018b), 96-well plates were inoculated with 100 μL of twofold serial dilutions of enrofloxacin, ampicillin and trimethoprim-sulfamethoxazole (Sigma-Aldrich, St. Louis, United States) in cation-adjusted Mueller-Hinton broth (CAMHB) to obtain final concentration ranges of 32-0.25 μg/mL, 256-2 μg/mL and 640-5 μg/ mL, respectively.Salmonella colonies were resuspended in sterile medium and the suspension turbidity was measured spectrophotometrically at an optical density of 600 nm.The bacterial suspension was then adjusted in CAMHB to 5 × 10 5 CFU per mL and vigorously vortexed.The plates were inoculated with 100 μL of bacterial suspension and incubated at 35°C for 20 h.Negative (sterile CAMHB) and positive (Salmonella isolate in CAMHB without antibiotics) controls were included in each plate.Escherichia coli ATCC 25922 was used as a quality control strain.MICs were interpreted according to CLSI VET08 clinical breakpoints for Enterobacteriaceae (Clinical and Laboratory Standards Institute (CLSI), 2018a): ampicillin MIC ≥32 μg/ mL resistant, 16 μg/mL intermediate, ≤8 μg/mL susceptible; enrofloxacin MIC ≥4 μg/mL resistant, 2-1 μg/mL intermediate, ≤0.5 μg/ mL susceptible; trimethoprim-sulfamethoxazole MIC ≥80 μg/mL resistant and ≤40 μg/mL susceptible.
Taxonomic identification PCR-RFLP was performed to provide a reliable taxonomic identification of 'black panthers'.All the ambiguous samples tested were successfully identified as Panthera pardus.Thus, the overall animals comprised 41 Panthera leo, 42 Panthera tigris, 15 Panthera pardus, 6 Puma concolor, 2 Acinonyx jubatus, 1 Panthera leo × Panthera tigris and 1 Leopardus pardalis.The number of faecal samples collected from each animal species is reported in Table from G. duodenalis retrieved in GenBank (AY775201, MG924430.1).
nellosis in Europe, in order S. Infantis, Enteritidis, Typhimurium, Derby and Newport (European Food Safety Authority (EFSA) & European F I G U R E 1 Percentage distribution of Salmonella enterica subsp.enterica serovariants isolated from faecal samples of the captive wild felids.

Table 1 .
The most prevalent serovariant was S. enterica subsp.en-Salmonellaentericasubsp.entericaserovariantsweresporadically detected.The percentage distribution is shown in Figure1.Forty representative isolates were selected to investigate the occurrence of antimicrobial resistances.The results are summarized in
TA B L E 2 Parasites detected from faecal samples of captive large felids sampled according to the housing site (circus, C1; zoos, Z2, Z3 and Z4).