Prevalence and Molecular Characterization of Cryptosporidium in Goats across Four Provincial Level Areas in China

This study assessed the prevalence, species and subtypes of Cryptosporidium in goats from Guangdong Province, Hubei Province, Shandong Province, and Shanghai City of China. Six hundred and four fecal samples were collected from twelve goat farms, and the overall infection rate was 11.4% (69/604). Goats infected with Cryptosporidium were found in eleven farms across four provincial areas, and the infection rate ranged from 2.9% (1/35) to 25.0% (9/36). Three Cryptosporidium species were identified. Cryptosporidium xiaoi (45/69, 65.2%) was the dominant species, followed by C. parvum (14/69, 20.3%) and C. ubiquitum (10/69, 14.5%). The infection rate of Cryptosporidium spp. was varied with host age and goat kids were more susceptible to be infected than adult goats. Subtyping C. parvum and C. ubiquitum positive samples revealed C. parvum subtype IIdA19G1 and C. ubiquitum subtype XIIa were the most common subtypes. Other C. parvum subtypes were detected as well, such as IIaA14G2R1, IIaA15G1R1, IIaA15G2R1 and IIaA17G2R1. All of these subtypes have also been detected in humans, suggesting goats may be a potential source of zoonotic cryptosporidiosis. This was the first report of C. parvum subtypes IIaA14G2R1, IIaA15G1R1 and IIaA17G2R1 infecting in goats and the first molecular identification of C. parvum and its subtypes in Chinese goats.


Introduction
Cryptosporidium is an intestinal protozoan parasite that reportedly infects humans and animals worldwide [1]. In neonatal ruminants, cryptosporidiosis is a leading cause of diarrhea and mortality, and causes farmers significant economic loss [2]. Numerous molecular biological techniques have detected Cryptosporidium species/genotype and subtypes and will improve our understanding of cryptosporidiosis transmission in man and animals [3].
Goats are economic resources in China. According to the National Bureau of Statistics, about 141.4 million goats were in China at the end of 2012 (http://data.stats.gov.cn/workspace/ index?m=hgnd), but few studies exist identifying the molecular epidemiology of cryptosporidiosis in goats. In one study, three species (C. ubiquitum, C. andersoni, and C. xiaoi) were found in Henan Province and one species (C. ubiquitum) was found in Chongqing City [20]. In another report, C. xiaoi and a novel Cryptosporidium genotype were found in goats in Qinghai Province [17]. Given the sparse data on prevalence and molecular characterization of Cryptosporidium spp. in goats in China, the present study aimed to supplement the genetic characterization of Cryptosporidium spp. in partial provinces of China, to investigate the subtypes of C. parvum and C. ubiquitum, and to evaluate the potential threat of Cryptosporidium spp. in goats to human health.

Ethics Statement
The protocol of this work was approved by the Animal Care and Use of Chinese Academy of Agricultural Sciences, and authorized by the Animal Ethical Committee of Shanghai Veterinary Research Institute. All the fecal samples obtained from goat farms were permitted by the owners of farms.

Sample Collection
Fresh fecal samples were collected from 12 goat farms across four provincial areas of China from November 2007 to September 2013. Two farms from Huizhou City (22u479N, 114u279E) are located in Guangdong Province (South China area), four from Gongan County (30u039N, 112u139E) in Hubei Province (Central China area), two from Tengzhou City (35u039N, 117u089E) in Shandong Province (East China area), and one from Chongming County (31u369N, 121u409E), one from Jinshan District (30u499N, 121u369E) and two from Fengxian District (30u559N, 121u229E) in Shanghai City (East China area). The weather across four provincial areas are different. Guangdong has a tropical and sub-tropical climate (warm and humid all year round), Hubei and Shanghai belongs to the sub-tropical monsoonal climates (hot, humid summers and generally mild winters), and Shandong has a warm-temperate monsoonal climate (hot, rainy summers and cold, dry winters) (http://en.wikipedia.org/wiki/ Category:Geography_in_China_by_province). The flocks from all the farms were similar, and the goats were kept within fenced areas at daytime and in house with wooden, slatted floors or on the cement floor at night time. Most goats drank from water channels (source from water supply) and took native grasses. Six hundred and four samples were obtained directly from the rectums or immediately after the animals defecated using sterile gloves. The samples were sealed in bags, taken back to the laboratory and processed within a week. The goats involved in this study had not previously been examined for Cryptosporidium infection.

DNA extraction, PCR amplification and sequence analysis
Genomic DNA was extracted from 300 mg of each faecal sample. All the samples were washed twice with sterile water and further purified using FastDNA SPIN Kit for Soil and the FastPrep Instrument (MP Biomedicals, Santa Ana, CA) in accordance with the manufacturer's instructions. A two-step nested PCR of small subunit (SSU) rRNA gene was used to amplify the DNA fragment, as described by Xiao et al. [23,24]. Positive and negative control samples were used for each PCR reaction. All secondary PCR products were sequenced as previously described [25]. Sequence analysis was guided using BLAST (http://blast.ncbi.nlm.nih.gov/Blast.cgi), and Cryptospo-ridium species/genotypes were identified preliminarily. A Neighbor-Joining (NJ) tree was constructed using/by MEGA 6 software (http://www.megasoftware.net/) to assess the genetic relationship between different Cryptosporidium species/genotype sequences [26]. The reference sequences of Cryptosporidium spp. were obtained from GenBank.

Subtype identification
For C. parvum, a ,400 bp fragment of the 60 kDa glycoprotein (gp60) gene was amplified by nested PCR as described by Sulaiman et al. [27]. For C. ubiquitum, a ,950 bp fragment of the gp60 gene was amplified by nested PCR as reported by Li et al. [28]. The second PCR products were purified using AxyPrep DNA Gel Extraction Kit (Axygen Scientific, Hangzhou, China) and sequenced as SSU rRNA gene. The Cryptosporidium subtypes were identified as described by Sulaiman et al. [27] and Li et al. [28].

Statistical analysis
Cryptosporidium spp. prevalence in different regions and ages was determined using IBM SPSS Statistics V21.0 for Windows (International Business Machines Corp, New York, USA). The differences were considered significant when p,0.05 by Pearson's Chi-Square test (x 2 test) analysis.
Eleven goat farms tested positively for Cryptosporidium. The highest infection rate was in Gongan-4 farm (25.0%, 9/36) of Hubei Province, while the Gongan-3 farm had no Cryptosporidium positive samples (Table 2). Among the eleven farms with Cryptosporidium infection, the infection rates ranged from 2.9% to 25.0% ( Table 2).
Infections mixing two species were found in four farms (Gongan-4, Huizhou-2, Fengxian-1 and Jinshan), and infection from all three species was found in Chongming. The unique partial SSU rRNA gene sequences of Cryptosporidium spp. were submitted to the GenBank database under accession numbers KM199743 to KM199759.
In addition to adult goats, C. xiaoi and C. parvum were found in three age groups. C. ubiquitum also found in three age groups except for pre-weaned kids. C. xiaoi (22/45, 48.9%) and C. ubiquitum (8/10, 80%) were mainly prevalence in post weaned kids, while C. parvum had a similar infection rates in three age groups ( Table 3).

Prevalence of Cryptosporidium subtypes
Twelve (85.7%) C. parvum positive samples were subtyped successfully based on the gp60 gene. Two subtypes, IIa and IId, were found. Four C. parvum IIa subtypes were found in 4 samples, including an IIaA14G2R1 from Gongan-1 farm and an IIaA15G1R1 from Gongan-4 farm of Hubei Province, and an IIaA15G2R1 and an IIaA17G2R1 from Jinshan farm of Shanghai City (Table 2). One IId subtype (IIdA19G1) was found in 8 samples from Chongming farm of Shanghai City ( Table 2). All C. ubiquitum positive samples (10/10) were subtyped successfully, and all of them belonged to subtype XIIa ( Table 2). The various sequences we acquired in this study were put in the GenBank database under accession numbers KM199736 to KM199742.
In this study, the prevalence and molecular characterization of Cryptosporidium in goats across four provincial level areas in China were detected with nested PCR technique. Eleven goat farms had Cryptosporidium infection, and the infection rates ranged from 2.9% (1/35) to 25.0% (9/36). Only one farm (Gongan-3) from Hubei Province was not found Cryptosporidium infection. We found different provinces had different infection rates, and the highest prevalence was detected in Shandong Province (18/100, 18.0%), followed by Hubei Province (13/111, 11.7%), Shanghai City (33/302, 10.9%), and Guangdong Province (5/91, 5.5%). The total prevalence of Cryptosporidium spp. in goats from 12 goat farms was 11.4% (69/604), which was higher than the 4.4% (10/228) infection rate in Papua New Guinea used the same detection method [22], and lower than the 17.9% (980/5468) infection rate in goats summarized by Wang et al [20]. In China, partial provinces reported Cryptosporidium infection in goats, but the reported infection rates varied greatly. For instance, the infection rate was 35.7% (15/42) by immunofluorescence test (IFT) and 4.8% (2/42) by PCR approach in Wulan County of Qinghai Province [17]. The infection rates were 27.5% (14/51) by IFT and 15.7% (8/51) by PCR in Qinghai Province [36]. A recent survey reported the percentage of Cryptosporidium infection was 2.75% (28/1017) in Henan Province and 6.45% (16/248) in Chongqing City by microscopical detection [20]. The difference in infection rates between our survey and previous surveys may be affected by host age, examination methods, feeding levels and the raising density of animals.
All positive samples were sequenced successfully, and three species were identified in this study. C. xiaoi was the most common species (45/69, 65.2%), followed by C. parvum (14/69, 20.3%) and C. ubiquitum (10/69, 14.5%). The results were similar to a recent study in Greece, in which same Cryptosporidium species were observed in goat kids [21]. The Cryptosporidium species presented differently depending on geographical location. C. xiaoi by itself was found in Shandong Province, C. xiaoi and C. parvum mixed were found in Hubei Province, C. xiaoi and C. ubiquitum mixed were found in Guangdong Province. All three species were found in Shanghai City. These results agree with previous surveys in China, which detected C. andersoni, C. ubiquitum and C. xiaoi in Henan Province, C. ubiquitum in Chongqing City [20], and C. xiaoi and a new genotype in Qinghai Province [17], but we did not find C. andersoni in goats for this survey.
C. xiaoi (previously identified as Cryptosporidium bovis-like genotype) was first reported in a goat in Qinghai Province of China [17] and subsequently identified in countries like France, Norway, Papua New Guinea, and Spain [10,15,16,22]. Recently, C. xiaoi was also found in Henan Province of China [20]. In this study, C. xiaoi as the dominant species was found in all the detected provincial areas and most of the detected goat farms. A previous French study found that C. xiaoi was the most common species in pre-weaned kids [15], and a recent study also reported a high prevalence of C. xiaoi infection in goat kids in Greece [21]. In China, C. xiaoi was also detected in sheep [37,38] and a yearling yak [39]. A recent study reported that C. xiaoi was identified in two HIV/AIDS patients in Ethiopia [40], suggesting that it pose a potential threat to human health.
Most previous surveys found that C. parvum was the dominant species in goats [5,8,9,11]. In this study, C. parvum was the second most prevalent species in goats. This agrees with a recent report from France in which C. xiaoi had a higher prevalence rate than C. parvum [15], and in Papua New Guinea where C. hominis was more prevalent than C. parvum [22]. To our knowledge, this was the first molecular identification of C. parvum infection in goats in China. C. ubiquitum (formerly known as Cryptosporidium cervine genotype) as a zoonotic species has been reported in humans worldwide and in many animals, including ruminants, primates, and rodents, and has also been found in water [1,27,41,42]. Few studies report it in goats. A recent survey in China have reported that C. ubiquitum was the predominant species in Chongqing City [20]. By contrast, we found C. ubiquitum was the least frequently detected species in goats. C. ubiquitum was found in only three goat farms (Huizhou-2, Chongming, and Fengxian-1). Interestingly, in Fengxian-1 farm of Shanghai City, C. ubiquitum was the dominant species.
Our survey, and previous surveys, revealed that young animals were more susceptible to infection than adult animals [29,43]. Four age groups in goats were identified in this study, and the infection rate was decreased significantly with the increase of host age. All three species were found in post weaned kids and yearling goats, and C. xiaoi and C. parvum were found in pre-weaned kids, while C. ubiquitum as a unique species was identified in an adult goat. C. xiaoi was the major species, followed by C. parvum and C. ubiquitum, and these results were different than in previous studies. Most previous reports found C. parvum as the dominant species in goat kids [5.8,10,11,14]. Other studies in France and Spain listed C. xiaoi as the most common species in goat kids [15,16], and C. ubiquitum as the unique species was found in periparturient goats in one flock in western France [18]. However, a recent study found that C. parvum and C. xiaoi existed in adult goats in Papua New Guinea [22], and C. ubiquitum and C. xiaoi were found in goats of all ages in China [20]. In this study, only one positive sample was detected in adult goats, it may be attributed to the small number of fecal samples. In future research, larger numbers of fresh samples of adult goats in these areas will be collected and detected to understand the relationship between the host age and Cryptosporidium species infection.
To date, only a few studies analyzed C. parvum subtypes in goats, and two subtype families (IIa and IId) were reported. The IIa family has been found in Belgium, Norway, and Papua New Guinea [5,10,22], and the IId family was only found in Europe (Belgium, Greece, Italy and Spain) [5,9,11,21]. Similar to previous studies, the IIa and IId families were also found in China in this survey, and IId subtype family (8/12, 66.7%) was the dominant subtype in goats. This agrees with previous surveys that said goats were more susceptible to infection C. parvum IId subtypes in Belgium (8/11, 72.7%), Greece (2/2, 100%) and Spain (17/17, 100%) [5,11,21]. The distribution of C. parvum subtypes had geographic differences. IIa subtypes were found in the Gongan-1 (IIaA14G2R1) and Gongan-4 (IIaA15G1R1) goat farms of Hubei Province and in Jinshan farm (IIaA15G2R1 and IIaA17G2R1) of Shanghai City. The subtype IId was identified only in Chongming farm (IIdA19G1) of Shanghai City. The subtype IIdA19G1 had been detected in goats in previous studies from Spain [11] and in HIV-positive patients in China [44] and Portugal [45]. IIdA19G1 also had been found in calves in Henan Province [46], but had never been found in goats in China. Subtype IIaA15G2R1 was the most common subtype in calves and humans [3,47]. It had been found in goats in Belgium [5] and Papua New Guinea [22]. In China, this subtype was also detected in lambs and yaks [38,48] but not goats. Subtype IIaA14G2R1 was mainly found in calves in Europe, including in Belgium [49], England [50], Germany [51], and the Netherlands [52]. It was also found in humans in Ethiopia [40]. In China, this subtype had been found in yaks [48]. Other subtypes, IIaA15G1R1 and IIaA17G2R1, have been reported in humans in many previous studies. For example, subtype IIaA15G1R1 had been found in humans in Australia [53], Egypt [54], Kuwait [26], and Slovenia [55]. Subtype IIaA17G2R1 had been reported in Australia [53,56,57], Canada [58], Ethiopia [40], the United Kingdom [59], and the United States [60]. The three IIa subtypes detected in this study, including IIaA14G2R1, IIaA15G1R1 and IIaA17G2R1, are the first reported in goats. Taken together, all the C. parvum subtypes have been detected previously in humans, suggesting that goats may be involved in zoonotic transmission of cryptosporidiosis.
Recent research about C. ubiquitum subtyping that targeted the gp60 gene was reported by Li et al. [28], and 6 subtype families (XIIa-XIIf) were identified. In this study, all the C. ubiquitum positive samples were identified as XIIa, in accordance with a recent survey that asserted XIIa was the unique subtype in goats in Henan Province and Chongqing City [20]. This subtype was also found in a goat from Algeria [28]. A recent study found C. ubiquitum subtype XIIa not only existed in domestic and wild ruminants, but also commonly seen in humans [28], which suggests this subtype may be a potential source of C. ubiquitum infection between animals and humans.
In summary, the prevalence and molecular characterizations of Cryptosporidium species and subtypes in goats from 12 farms across four different provincial level regions of China indicate that Cryptosporidium spp. are common infections in goats in China. The overall infection rate was 11.4% and the highest infection rate was found in pre-weaned kids. C. xiaoi, five subtypes of C. parvum, and one subtype of C. ubiquitum were found in goats. To our knowledge, this was the first report of C. parvum subtypes IIaA14G2R1, IIaA15G1R1 and IIaA17G2R1 infecting in goats and the first molecular identification of C. parvum and its subtypes in Chinese goats. The zoonotic C. parvum subtype IIdA19G1 and C. ubiquitum subtype XIIa were the dominant subtypes in the present study, suggesting that monitoring goat populations for the presence of Cryptosporidium is important to public health.