Biosurveillance of Selected Pathogens with Zoonotic Potential in a Zoo

Monitoring of infectious diseases is one of the most important pillars of preventive medicine in zoos. Screening for parasitic and bacterial infections is important to keep animals and equipment safe from pathogens that may pose a risk to animal and human health. Zoos usually contain many different animal species living in proximity with people and wild animals. As an epidemiological probe, 188 animals (122 mammals, 65 birds, and one reptile) from a zoo in Slovenia were examined for selected pathogens. Antibodies to Toxoplasma gondii and Neospora caninum were detected by ELISA in 38% (46/122) and 3% (4/122) of mammals, and in 0% (0/64) and 2% (1/57) of birds, respectively; the reptile (0/1) was negative. A statistically significant difference in T. gondii prevalence was found in Carnivora compared to Cetartiodactyla and primate antibodies to Encephalitozoon cuniculi were detected by IFAT in 44% (52/118) of mammals and 20% (11/56) of birds, respectively; the reptile (0/1) was negative. Herbivores had a higher chance of being infected with E. cuniculi compared to omnivores. Antibodies to Chlamydia abortus and Coxiella burnetii were not detected in any of the 74 tested zoo animals. The sera of 39 wild rodents found in the zoo were also examined; they were negative for all three parasites. The parasite T. gondii was detected by PCR in the tissue of two mute swans (Cygnus olor), three eastern house mice (Mus musculus), one yellow-necked field mouse (Apodemus flavicollis), and one striped field mouse (A. agrarius). Positive samples were genotyped by a single multiplex PCR assay using 15 microsatellite markers; one sample from a mute swan was characterized as type II. This micro-epidemiological study offers a better understanding of pathogens in zoo animals and an understanding of the role of zoos in biosurveillance.


Introduction
The role of zoos in biosurveillance has been demonstrated in the past [1]. Many different animal species usually live in a small area and in proximity to keepers, visitors, and other wildlife. In some situations, infectious pressure might overcome the immune defense of zoo animals, and pathogens might spread across various animal species. T. gondii and N. caninum are protozoal parasites with the ability to cause disease in a wide spectrum of animals [2]. Many species living in captivity (New World monkeys, lemurs, Pallas cats, marsupials, etc.) have died of clinical toxoplasmosis. There is also potential risk of exposure of children and the elderly to T. gondii oocysts excreted by the cats in zoos [3,4]. In contrast to T. gondii, N. caninum is not considered to be zoonotic, although there is serologic evidence of human exposure, primarily in immunocompromised people [5,6]. N. caninum was studied in recent years in terms of its veterinary importance, especially in canids, and economic losses in cattle due to reproductive problems [7,8]. Reproductive problems associated with N. caninum infection were reported also in southern white rhinoceros (Ceratotherium simum simum) from an Australian zoo [9]. Encephalitozoon cuniculi belongs to microsporidia group and is one of the three pathogenic species from the genus Encephalitozoon. Differentiation of microsporidiosis from other diseases in captive animals is difficult because of nonspecific clinical signs. This disease could remain unrecognized or misdiagnosed. In exotic animals, E. cuniculi usually does not cause any symptoms; nevertheless, some fatal occurrences have been recorded. [10][11][12][13][14]. Chlamydia abortus and Coxiella burnetii are obligate intracellular bacteria. The Ch. abortus is a zoonotic pathogen and causes reproductive failure, especially in sheep, goats, and cattle. This etiologic agent was also isolated in wild ruminants and zoo animals as a cause of abortions [15,16]. The Co. burnetii is the widespread zoonotic initiator of Q fever causing illnesses in human and livestock. Domestic and wild animals, birds, amphibians, and arthropods play the role of reservoirs of infection [17]. Although infection in livestock is mainly asymptomatic, it can lead to abortion, reproductive disorder, weak offspring, or infertility [18]. This disease was also diagnosed as a cause of abortion in zoo animals and was isolated from zoo and wildlife ungulates in Slovakia, Spain, and Portugal [19][20][21][22].
In this context, infectious diseases monitoring is an extremely important part of preventive veterinary medicine in zoos. A well-designed vaccination program and regular coprological examinations, including flotation and sedimentation techniques, are obligatory in order to provide good health care for animals in the zoo. A detailed screening plan for selected pathogens based on the current epidemiological situation is of imminent importance for the safety of animals and employees, as well as visitors. The possibility of discovering infectious threats, such as toxoplasmosis, neosporosis, chlamydiosis, and encephalitozoonosis, makes zoos suitable epidemiological stations. The aim of this study was to monitor selected pathogens at the Ljubljana Zoo in Slovenia and examine its role as a sentinel in biosurveillance.
All the birds tested were negative for antibodies to T. gondii (0/64) and Ch. abortus (0/26). Antibodies to N. caninum were found in one animal (1/57), in a 14-year-old female blue-and-yellow macaw (Ara ararauna), and antibodies to E. cuniculi were found in 20% (11/56) of animals, in two 8-and 15-year-old male mute swans (Cygnus olor) and five female and four male helmeted guineafowls (Numida meleagris), 6 months to 5 years old. The presence of both antibodies to N. caninum and E. cuniculi was not proved.
One reptile was negative for T. gondii and N. caninum antibodies; E. cuniculi was not tested in this animal.
Antibodies to T. gondii, N. caninum, E. cuniculi, and Co. burnetii were not found in any of the 39 wild rodents; antibodies to Ch. abortus were not examined.
The T. gondii DNA was detected by PCR in the tissue of seven of 21 (33%) animals that died at the zoo or were trapped during deratization. The positive tissues were livers of two mute swans and brains of three eastern house mice (Mus musculus), one yellow-necked field mouse (Apodemus flavicollis), and one striped field mouse (A. agrarius). The sample from a mute swan (Cygnus olor) was successfully genotyped and characterized as type II. The parasites N. caninum and E. cuniculi were not found by PCR in any of the tissue samples.

Discussion
The main source of parasitic infection in zoo animals is not well known. At the Ljubljana Zoo, as in other European zoos, feeding carnivores with raw meat that is potentially infected with T. gondii tissue cysts is a common practice. Infected felids can spread T. gondii oocysts in their environment and thus contaminate food, bedding, and water. In some zoo felids (Otocolobus manul), toxoplasmosis with a fatal outcome has been reported [23]. Carnivores can also be infected by hunting some small prey species with free access to the zoo, such as small wild rodents and birds. Another source of infection might be the presence of feral cats in the zoo, which can spread oocysts in their feces to zoo enclosures. In our study, T. gondii was detected by PCR in tissues of four wild rodents and in one eastern house mouse that died. The eastern house mouse came from a specific pathogen-free breeding compound; the route of infection of this animal remains unknown. One possibility is food contaminated with sporulated T. gondii oocysts due to improper storage of prepared pellets.
Fatal toxoplasmosis was described in six tammar wallabies (Macropus eugenii) in the Budapest Zoo and Botanical Garden between 2006 and 2010 [24]. Wallabies are known to be susceptible to T. gondii infection, and feral cats are suggested as the main source of the parasite in the zoo environment. In our study, a high prevalence of T. gondii was found in Camelids and Diprotodontia, which could be related to contamination of food, such as hay, or of the substrate material in their enclosures, which is mostly sand. A high risk of contamination of this substrate with the feces of feral cats, as well as difficulties in keeping stray cats outside the zoo (it is located in the city), increase the possibility of infection with T. gondii oocysts. New World primates are highly susceptible to clinical toxoplasmosis, and the infection is often fatal with various pathological manifestations [25]. High prevalence was previously recorded in primates such as the red-faced spider monkey (Ateles paniscus) and tufted capuchin (Cebus apella) in a zoo in Brazil, at 67% [26], and in a Barbary macaque (M. sylvanus), gorilla (Gorilla gorilla), chimpanzee (Pan troglodytes), and orangutan (Pongo pygmaeus) in the Czech Republic, at 45% [27]. In our study, T. gondii antibodies were found in 25% of primates (chimpanzee, Pan troglodytes). However, no clinical symptoms consistent with this infection have been observed. Ingestion of sporulated T. gondii oocysts in contaminated feed is presumed to be the main source of infection. Carefully washing vegetables and fruits as well as regularly cleaning the kitchen may reduce the risk of infection in this group of animals. Toxoplasmosis has been recorded in zoo birds; for example, in canaries and black-winged lories (Eos cyanogenia) [28]. In our study, antibodies to T. gondii were not found in any of the birds tested; however, two mute swans died and T. gondii was detected in their livers by PCR. These samples were genotyped and characterized as type II, which is known to be highly predominant in humans and animals in Europe and North America [29,30]. Water contaminated with T. gondii oocysts is one possible source of the infection in this case.
Antibodies to N. caninum were detected in a small number of animals. Of the mammals, only two domestic sheep and two European fallow deer were positive. The role of ruminants in the life cycle of this parasitic species is very well known and there are many studies describing the clinical signs and laboratory diagnosis of Neospora-positive captive and wild ruminants including those from zoos [8,31]. There were no clinical signs of N. caninum infection in the anamnestic history of positive ruminants from our study. In birds from Ljubljana zoo, antibodies to N. caninum were found in one female blueand-yellow macaw (Ara ararauna). In psittacine birds that died from unrelated clinical conditions, N. caninum tissue cysts were demonstrated in the muscles of red-and-green macaw (Ara chloropterus) and blue-fronted Amazon parrots (Amazona aestiva) from Brazil using the imunohistochemical method [32]. The role of birds in the life cycle of N. caninum is still being investigated; however, the results of several experimental studies or studies on wild and captive birds suggest that they can serve as potential reservoirs of Neospora infections, especially as prey for wild canines, as well as a mechanical vector of N. caninum oocyst [8,33]. Birds can thus be an important source of infection for animals living in zoos in outdoor exhibitions, as well as wild canines (stray dogs, foxes, etc.) which can serve as potential definitive hosts of N. caninum in the area of the zoo.
E. cuniculi is a very important parasite of rabbits throughout Europe [34]. This is in accordance with results from our study because we found the highest prevalence of E. cuniculi antibodies in Lagomorpha (83%) in European rabbits (Oryctolagus cuniculus). Placentitis, premature birth, and perinatal death has been described in connection with E. cuniculi infection in an alpaca (Vicugna pacos) [35]. In our study, a high prevalence of E. cuniculi antibodies was found in Bovidae (81%) and Camelidae (67%). Antibodies to E. cuniculi were found in an alpaca (Vicugna pacos) and a guanaco (Lama guanicoe). One E. cuniculi-seropositive alpaca died, but the parasite was not detected by PCR in its tissue.
Clinical disease and positive isolation of E. cuniculi was reported in captive emperor tamarins (Saguinus imperator) and cotton-top tamarins (Saguinus oedipus), which seem to be highly susceptible to infection [10,11]. In our study, chimpanzees, lemurs, and gibbons were negative. Herbivores had a higher chance of being infected compared to carnivores, which could suggest the presence of E. cuniculi spores in the environment as a main source of infection. E. cuniculi infection in birds has no clinical importance, but some avian species could serve as reservoirs for these microsporidia [36]. In our study, antibodies to E. cuniculi were found in mute swans and helmeted guineafowls (Numida meleagris).
Chlamydia abortus and Coxiella burnetii are pathogens that are well known to cause abortions as well as various health problems in zoo animals [22]. The negative results obtained in our study may have been due to limited exposure, good preventive veterinary care, and suitable disinfection plans.
Zoos are institutions open to the public. A large variety of different animal speciesusually with a high density of animals-open enclosures, the presence of stray cats and wild rodents, and close contact with people are important factors for transmission of diseases, often with anthropo-zoonotic potential, which raises public health concerns. Therefore, monitoring transmissible diseases in zoo animals and understanding their dynamics is of great importance and is an inseparable part of preventive health care in zoos.

Materials and Methods
The zoo is located on the outskirts of Ljubljana (coordinates 46 • 3 9.25 N, 14 • 28 20.08 E) and covers an area of 6 hectares with a large collection of exotic mammals, birds, and reptiles. Blood samples (n = 188) were collected from the most appropriate vein based on the animal species for 122 mammals, 65 birds, and 1 reptile in the years 2014 and 2015 (Table 1), and the serum was stored at −20 • C until assay. Tissue samples (brains or liver) of ten animals that died at the Ljubljana Zoo in 2015 were also examined. These included two mute swans (Cygnus olor), one black stork (Coconia nigra), two brown rats (Rattus norvegicus var. alba), one fat-tailed gerbil (Pachyuromys duprasi), one alpaca (Vicugna pacos), two Japanese quail (Coturnix japonica), and one eastern house mouse (Mus musculus) used as food for zoo animals. In general, samples from majority animals were collected during the preventive clinical health checks or elective surgical procedures for preventive purposes. Animals were clinically healthy with no obvious signs of any disease. Some of the samples were collected from the animals with clinical signs, where obtaining of samples was a necessary part of diagnostic procedure. In some zoo animals, the age was known, especially if they were born in the zoo. This varied across the species, from young to very old (even older than would be possible in the wild). In other animals, only a rough estimation of the age was made. Animals were not vaccinated against any of pathogens tested as these etiological agents were not part of the standard operation policy for vaccination in this zoo.
As a part of the deratization program at the Ljubljana Zoo, sera from 39 wild rodents (n = 39) and their tissue (brain or liver) (n = 11) were collected and used for assays. The rodents included 28 eastern house mice (Mus musculus), four brown rats (Rattus norvegicus), four yellow-necked mice (Apodemus flavicollis), two bank voles (Clethrionomys glareolus), and one striped field mouse (A. agrarius). All the samples used in this study were either blood samples collected during surgeries and preventive annual routines where blood was used for other reasons, or tissues from dead animals and rodents collected during the deratization program. For molecular analyses, the examined tissues were selected according to parasite distribution in tissues (the liver was used for detection of E. cuniculi DNA and the brain was used for detection of T. gondii and N. caninum DNA). An enzyme-linked immunosorbent assay (ELISA) was used to detect antibodies to T. gondii, N. caninum, Co. burnetii, and Ch. abortus with the ELISA kits ID Screen T. gondii Indirect Multi-Species, ID Screen N. caninum Indirect Multi-Species, ID Screen Q Fever Indirect Multi-Species, and ID Screen Chlamydia abortus Indirect Multi-Species (IDvet, Grabels, France), respectively. Samples with sample-to-positive ratio (S/P) for antibody ELISA of S/P ≥ 60% and S/P ≥ 50% were classified as positive for Ch. abortus and for the other infections mentioned above, respectively. Antibodies to E. cuniculi were determined with an indirect immunofluorescent antibody test using MegaScreen Fluoencephalitozoon (Megacor Diagnostic, Hörbranz, Austria) with E. cuniculi antigen. The specific conjugates were the following: anti-bovine, anti-goat, and anti-sheep IgG (VMRD, Pullman, Chicago, USA) for Bovidae; anti-camel IgG (VMRD) for Camelidae; anti-deer IgG (KPL Inc. Gaithersburg, MD, USA) for Cervidae; anti-horse IgG (VMRD) for Equidae; anti-cat IgG (Sigma Aldrich, St. Louis, MO, USA) for Felidae and Ursidae; antidog IgG (Sigma Aldrich) for Canidae; anti-pig IgG (Sigma Aldrich) for Suidae; anti-chicken (Sigma Aldrich) for Galiiformes; anti-duck (KPL Inc. Gaithersburg) for Anseriformes; and anti-budgerigar (KPL Inc. Gaithersburg) for Psittacidae, Accipitriformes, Ciconiiformes, Pelacaniformes, Rheiformes, Strigiformes, and Struthioniformes. Sera were diluted with phosphate-buffered saline twofold, starting with 1:50; samples with a titer ≥ 50 were marked as positive.
The DNA was isolated from the brain or liver by the QIAamp DNA Mini Kit (Qiagen, Courtaboeuf, France); in the case of E. cuniculi, the tissue (liver) was first homogenized using a tissue homogenizer. Detection of T. gondii DNA was done by PCR amplification of the TGR1E sequence [37]. The prevalence was statistically analyzed with Pearson's chi-squared test for independence using Statistica Cz 12 [41] or with the Monte Carlo method using IBM SPSS Statistics 20. The null hypothesis that seroprevalence in mammals does not differ by orders, families, diet, and sex was tested. The differences were considered statistically significant when the p-value was ≤ 0.05. In the case of a statistically significant difference, the Scheffé multiple comparison method (Statistica Cz 12) was subsequently applied.

Data Availability Statement:
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.