Molecular identification of Cryptosporidium spp. in seagulls, pigeons, dogs, and cats in Thailand.

Zoonotic Cryptosporidium spp., particularly C. meleagridis, C. canis, and C. felis, are enteric protozoa responsible for major public health concerns around the world. To determine the spread of this parasite in Thailand, we conducted molecular identification of Cryptosporidium spp. from animal samples around the country, by collecting and investigating the feces of seagulls (Chroicocephalus brunnicephalus and Chroicocephalus ridibundus), domestic pigeons (Columba livia domestica), dogs, and cats. Seagull and pigeon samples were collected at the seaside and on the riverside to evaluate their potential for waterborne transmission. Ten pigeon samples were combined into one set, and a total of seven sets were collected. Seventy seagull samples were combined into one set, and a total of 13 sets were collected. In addition, 111 dog samples were collected from cattle farms, and 95 dog and 80 cat samples were collected from a temple. We identified C. meleagridis in pigeons, Cryptosporidium avian genotype III in seagulls, C. canis in dogs, and C. felis in cats. In the temple, the prevalence was 2.1% (2/95) for dogs and 2.5% (2/80) for cats. No Cryptosporidium was found in dog samples from cattle farms. These are the first findings of C. meleagridis in domestic pigeons, and Cryptosporidium avian genotype III in seagulls. Our study invites further molecular epidemiological investigations of Cryptosporidium in these animals and their environment to evaluate the public health risk in Thailand. K. Koompapong et al., published by EDP Sciences, 2014

Introduction

Cryptosporidium, a common intestinal parasite of humans and animals, has a diverse global distribution and 30 species with >50 genotypes have been identified [47, 60]. In Europe, C. parvum and C. hominis are responsible for >95% of human cryptosporidiosis [9, 25, 51]. In the USA, Australia, and Japan, C. hominis is more prevalent than C. parvum [29, 59, 62]. C. meleagridis, C. felis, and C. canis, predominant species in birds, cats, and dogs, respectively, are generally less common than C. hominis and C. parvum [60]. In Thailand, cryptosporidiosis prevalence rates between 2.5 and 25% have been reported in HIV-infected patients in urban areas [53]. Particularly, C. meleagridis, C. felis, and C. canis have been found and are responsible for 35.2% of all Cryptosporidium infections in Thai HIV patients [11]. The risk of non-parvum zoonotic Cryptosporidium infection is considered high in Thailand, at least in immunodepressed patients.

To date, Cryptosporidium has been studied in dogs [3, 31], cattle [18, 19, 35], mussels [54], and chickens in Thailand, [20], but no molecular epidemiological investigations of Cryptosporidium in animals other than ruminants have been conducted. The main objective of our study was to identify Cryptosporidium in animals, especially focusing on birds, dogs, and cats, by a molecular method. We selected seagulls and domestic pigeons, common migratory and domestic birds that live near bodies of water, making them potential sources of water contamination. Many people visit seaside piers and riverside areas in Thailand for relaxation, and these areas often act as sanctuaries for birds since bird-feeding is a favored activity among both visitors and local people. In these areas, birds gather at precise times, waiting for people to feed them. They tend to be unafraid of human presence, and this proximity risks making these places important transmission zones for zoonotic diseases. For similar reasons, dogs in cattle farms, as well as dogs and cats in a temple were selected for investigation. Stray dogs and cats are often cared for at temples in Thailand due to religious beliefs, and temples have become known as endemic places for intestinal parasitic infections in animals living there [56].

Materials and methods

Study sites and sample collection

Seagull samples were collected at Bang Poo Nature Reserve Pier, in Samut Prakan Province, central Thailand. Most seagulls were brown-headed gulls (Chroicocephalus brunnicephalus) and black-headed gulls (Chroicocephalus ridibundus). These birds migrate from China in October–May each year, and the samples were collected in both 2010 and 2011. Due to the small amount of feces per dropping, fecal samples from 70 seagulls were combined into one set, and a total of 13 such sets were collected.

Samples from domestic pigeons (Columba livia domestica) were collected in September 2012 from three locations: (1) near the pier in Wat Rakang Kositaram; (2) Brahmin Swing; (3) the Pramane Ground. All locations are in Bangkok, central Thailand. Fecal samples from 10 pigeons were combined into one set, and a total of seven sets were collected. All bird samples were collected immediately when the feces dropped to the ground.

One hundred eleven dog samples were collected from 105 dairy cattle farms in central Thailand (Nakhon Pathom Province: Mueang Nakhon Pathom District; Ratchaburi Province: Photharam District, Ban Pong District; Lopburi Province: Phatthana Nikhom District; Sa Kaeo Province: Watthana Nakhon District, Mueang Sa Kaeo District, Wang Sombun District) in January–April 2012. Most of the dogs were adult. They appeared to be healthy and well fed by their owners. Dog samples were collected per rectum by experienced veterinarians. In addition, 95 samples from dogs and 80 from cats were collected from a temple in central Thailand (Ban Na District, Nakhon Nayok Province) in October 2012. About 1,000 dogs and 500 cats lived around the temple area, fed by monks and volunteers. Most of the dogs and cats were adult. Some animals were kept in cages, while others were free to roam. Dog and cat fecal samples were collected at random from the grounds. Samples from all animals were kept in dry tubes in cool conditions during transportation, and preserved at −20 °C for DNA extraction. This study was approved by the Animal Care and Use Committee, Faculty of Tropical Medicine, Mahidol University, Thailand.

Molecular analysis

DNA was extracted from the samples using a commercially available DNA extraction kit (PSP Spin Stool DNA Kit, Stratec Inc., Germany) according to the manufacturer’s instructions. This kit contains optimized essential washing conditions to remove inhibitors very efficiently. Fragments of SSU rRNA (830 bp) were amplified by PCR, using primers and protocols described previously [61]. The PCR products were subjected to electrophoresis in 2% agarose gel and visualized by ethidium bromide staining. All amplified products were sequenced in both directions using the secondary PCR primers on an ABI 3730xl DNA Analyzer (Applied Biosystems). The Cryptosporidium species and genotypes from each specimen were confirmed by homology of the sequenced PCR products to the sequence published in GenBank.

Results

The numbers of positive samples examined for Cryptosporidium for each collection site and the Cryptosporidium species/genotypes determined by PCR analysis of the SSU rRNA gene are summarized in Table 1. Two sets from seagulls, and one set from pigeons, were found positive for Cryptosporidium spp. with 100% homology with Cryptosporidium avian genotype III and C. meleagridis in the published sequences in GenBank (AB694729 and KF701463). Cryptosporidium was not identified in the dog samples from the cattle farms but 2/95 (2.1%) dog samples and 2/80 (2.5%) cat samples from the temple were found positive for Cryptosporidium. Genotyping analysis of the PCR-positive samples identified C. canis and C. felis, respectively, with 100% homology with the published sequences in GenBank (FJ233035 and JQ413437).

Table 1.

Number of positive samples examined for Cryptosporidium for each collection site and the Cryptosporidium species/genotypes determined by PCR analysis of the SSU rRNA gene.

Type of animal	Collection site	Sample size	Cryptosporidium-positive (%)	Cryptosporidium species/genotype
Pigeons	Wat Rakang Kositaram, Bangkok	4 sets*	1 (25%)	C. meleagridis
	Brahmin Swing, Bangkok	2 sets*	0
	The Pramane Ground, Bangkok	1 set*	0
Seagulls	Bang Poo Nature Reserve Pier, Samut Prakan	13 sets**	2 (15.3%)	C. avian genotype III
Dogs	Nakhon Pathom	3	0
	Ratchaburi	27	0
	Lopburi	33	0
	Sa Kaeo	48	0
	Nakhon Nayok	95	2 (2.1%)	C. canis
Cats	Nakhon Nayok	80	2 (2.5%)	C. felis

1 set = 10 fecal samples;

1 set = 70 fecal samples.

Discussion

Some Cryptosporidium species and genotypes are more associated with human illness than others, and some may be related to specific pathogenicity in human cryptosporidiosis [47]. Previously in Thailand, Cryptosporidium infections have been reported in food (12.5% in green mussels) [54], and the environment (12.7% in the waters of southwest coastal areas of Thailand) [52]. Unfortunately, species/genotype identification was not possible in previous studies in Thailand due to the use of the indirect fluorescent-antibody method and because no species-specific monoclonal antibodies were available. Otherwise, more detailed information on genotyping, particularly for human-adapted species, would allow more precise public health risk assessments for cryptosporidiosis in the future [53]. Similarly, our previous study using the nested PCR technique targeting the SSU rRNA gene revealed that 11% of river- and 6% of sea-water samples in central Thailand were contaminated with several Cryptosporidium species, including C. parvum, C. meleagridis and C. serpentis. The highest river contamination levels occurred during the rainy season, and the highest sea-water levels corresponded with the presence of migratory seagulls, indicating that runoff water carries the parasite into the rivers/sea [23]. Therefore, it is important to investigate the influence of domestic animals as potential transmission reservoirs for Cryptosporidium, and their impact on public health.

The domestic pigeon is one of the most common birds around the world. Herein, C. meleagridis was found for the first time in feces of domestic pigeons. To date, three avian Cryptosporidium species (C. meleagridis, C. baileyi, and C. galli) and 11 genotypes (avian genotypes I–V, black duck genotype, Eurasian woodcock genotype, and goose genotypes I–IV) have been reported [10, 58, 59]. Among them, C. meleagridis is pathogenic, causing diarrhea in both humans and birds [8, 41, 60]. As shown in Table 2, C. meleagridis has been identified globally in livestock, such as chickens, hens, and turkeys [2, 5, 13, 17, 33, 46, 48, 50, 55]. It has also been frequently reported in a variety of pet birds in Japan and China [1, 42]. A few studies have also found C. meleagridis in wild and domestic free-living birds [24, 30, 37].

Table 2.

Avian host and country of identification of C. meleagridis and Cryptosporidium avian genotype III, from published records.

	Avian host	Country	References
C. meleagridis	Chicken and hen (Gallus gallus domesticus)	USA	[2, 5, 13, 17, 33, 46, 50]
		Sweden
		Algeria
		Brazil
		Tunisia
	Turkey (Meleagris gallopavo)	USA	[5, 48, 55]
		Italy
		Algeria
	Ring-necked parrot (Psittacula krameri)	Australia	[30]
	Cockatiel (Nymphicus hollandicus)	Japan	[1]
	Red-legged partridge (Alectoris rufa)	China	[42]
		Spain	[37]
	Rose-ringed parakeet (Psittacula krameri)		[42]
	Bohemian waxwing (Bombycilla garrulus)		[42]
	Rufous turtle dove (Streptopelia orientalis)		[42]
	Fan-tailed pigeon (C. l. domestica)		[42]
	Scale quail (Callipepla squamata)		[42]
	Domestic pigeon (C. l. domestica)	Thailand	This study
Cryptosporidium avian genotype III	Cockatiel (N. hollandicus)	Australia	[33, 34, 42]
		Brazil
		China
	Galah (Eolophus roseicapilla)	Australia	[34]
	Sun conure/parakeet (Aratinga solstitialis)	Australia	[34]
	Canada goose (Branta canadensis)	USA	[21]
	Red-billed blue magpie (Urocissa erythrorhyncha)	China	[42]
	Peach-faced lovebird (Agapornis roseicollis)	Brazil	[33]
	Brown-headed gull (C. brunnicephalus)	Thailand	This study
	Black-headed gull (C. ridibundus)

Several recent studies report occurrences of Cryptosporidium in urban pigeons that are infectious to humans [15, 36, 42, 43]. Pigeons living in urban parks may be a source of zoonotic cryptosporidiosis, affecting primarily immunocompromised people, children, and the elderly [42]. We chose to collect domestic pigeon and seagull samples for this study in locations where many tourists gather to feed the birds. Our results indicate the need for further molecular epidemiological investigations of Cryptosporidium in domestic birds and their handlers, including pigeons, to evaluate the risk they pose to humans. Cryptosporidium are also well known for waterborne transmission as they are commonly found in wastewater [7, 24]. Our previous surveillance of river- and sea-water identified several Cryptosporidium species, including C. parvum, C. meleagridis, and C. serpentis [23]. The habitats of the pigeons in our study were very close to the pier and riverside, so they may cause water contamination.

Cryptosporidium avian genotype III was detected in seagull samples (Chroicocephalus brunnicephalus and Chroicocephalus ridibundus). Two previous studies have reported Cryptosporidium in seagulls, but this is the first study to identify Cryptosporidium genotypes in seagulls using a molecular method [40, 49]. Cryptosporidium avian genotype III has been detected in a variety of birds around the world (Table 2), such as the cockatiel (Nymphicus hollandicus), galah (Eolophus roseicapilla), sun conure/parakeet (Aratinga solstitialis), Canada goose (Branta canadensis), red-billed blue magpie (Urocissa erythrorhyncha), and peach-faced lovebird (Agapornis roseicollis) [21, 33, 34, 42], but has not been identified in livestock or other migratory birds. Because seagulls migrate long distances, they may be responsible for transmitting the parasite to other countries.

The prevalence of zoonotic Cryptosporidium spp. varies in different regions of the world. As shown in Table 3, in dogs, C. canis was the most prevalent species in Australia [39], C. parvum in Italy [12], and C. muris in the USA [26]. Cryptosporidium prevalence in dogs varies – high infection rates (>10%) have been reported in the USA and Norway [16, 26], with lower prevalence (<2%) in the UK, Brazil, and Australia [6, 32, 39]. C. felis is the predominant species in cats [4, 38, 45], while Cryptosporidium species tend to differ by study area. In cats, rates are generally higher than dogs; infection rates > 10% have been reported in Colombia, Italy, and the USA [4, 44, 45], and are reportedly low in the UK [14, 57]. In the present study, 2 out of 95 (2.1%) dog and 2 out of 80 (2.5%) cat fecal samples were positive for Cryptosporidium. The PCR technique, capable of detecting low levels of Cryptosporidium infection, was used in this study because Cryptosporidium can be difficult to detect using conventional microscopy, as found in previous reports [27, 28]. However, the prevalence of Cryptosporidium in dogs and cats was low. Given the low prevalence of this parasite found in our study, we suggest that dogs and cats do not pose a serious Cryptosporidium infection risk to humans in central Thailand. We expected to detect C. parvum in the dog samples from cattle farms, due to their proximity to cattle; the prevalence of Cryptosporidium in cattle in Thailand has been reported to be 9.5–13.0% [22, 35]. However, no C. parvum was found in the dog samples from the cattle farms. The reason for this negative result is unknown; specific C. parvum subtypes may only infect dogs. Further epidemiological and subtype studies of C. parvum in dogs are required.

Table 3.

Country and Cryptosporidium species for dogs and cats, from published records.

Animal	Country	Cryptosporidium species (number of identified samples)	References
Dog
	Australia	C. canis (4)	[39]
	Italy	C. canis (1) and C. parvum (7)	[12]
	Thailand	C. canis (2)	This study
	USA	C. muris (6)	[26]
Cat
	Australia	C. felis (18)	[39]
	Colombia	C. felis (5) and C. muris (1)	[45]
	Thailand	C. felis (2)	This study
	USA	C. felis (12)	[4]

Currently, no molecular tools are available to evaluate the subtype characteristics of C. canis or C. felis; therefore, the host specificity of these species has not been fully explored. In contrast, subtype analyses are available for C. parvum, and the variety of host specificity has been reported; some subtypes (IIa, IId) are zoonotic, while another (IIc) is anthroponotic [58]. Unfortunately, C. parvum subtypes have not been reported for dogs.

This study had some limitations. First, the sample size was not large enough to analyze the species/genotype characteristics in the region fully. Second, the birds’ fecal samples were combined into sets (10 samples = 1 set for pigeons, 70 samples = 1 set for seagulls), so that the actual prevalence could not be precisely determined. In view of the above findings, further studies are necessary as the number of seagulls, pigeons, dogs, and cats is very large and their proximity to humans makes contamination very likely.

In conclusion, we identified C. meleagridis in pigeons, Cryptosporidium avian genotype III in seagulls, C. canis in dogs, and C. felis in cats. Our study indicates that further molecular epidemiological investigations of Cryptosporidium in animals, especially pigeons, are necessary to evaluate their possible role as reservoir hosts, and the potential risk they pose to humans.