Revisiting the infectivity and pathogenicity of Cryptosporidium avium provides new information on parasitic sites within the host

Background Cryptosporidium spp. are protozoans that cause diarrheal illness in humans and animals, including birds, worldwide. The present study was aimed to revisit the infectivity and pathogenicity of C. avium, recently considered to be a valid avian-infecting species of Cryptosporidium, and foster further understanding of its biological characteristics. Results Results showed that no Cryptosporidium oocysts were detected in the feces of experimentally inoculated BALB/c mice, Mongolian gerbils, quail or budgerigars within 30 days post-infection (dpi). Oocysts were first detected in feces of 3-day-old and 40-day-old hens at 8 and 9 dpi, respectively. In ducks infected with C. avium, oocysts were first detected at 9 dpi. Oocysts of infected animals were studied using a nested-polymerase chain reaction (PCR) technique for the SSU rRNA gene, actin gene, HSP70 gene and Cryptosporidium oocyst wall protein gene (COWP) detection. Restriction fragment length polymorphism (RFLP), using SspI and VspI restriction enzymes, was carried out to genotype the species and obtained amplification products were sequenced. Cryptosporidium developmental stages were found in the longitudinal plica of the bursa fabricii (BF) of hens, with high levels observed in histological sections and scanning electron microscopy. No pathological changes were observed. Conclusions These findings indicate that the bursa fabricii may be the primary site of C. avium infection. More biological data are needed to support the establishment of new species and contribute to the taxonomy of Cryptosporidium. Electronic supplementary material The online version of this article (10.1186/s13071-018-3088-x) contains supplementary material, which is available to authorized users.

Cryptosporidiosis in birds has a wide spectrum of clinical signs, varying from asymptomatic to serious infection to death, and has been mainly associated with high morbidity and mortality in poultry [15,16]. Cryptosporidium baileyi is the most commonly-reported species in birds, with clinical signs including dyspnea, coughing, sneezing and depression [17]. Infection with C. galli primarily causes diarrhea, chronic apathy, weight loss and high mortality [4]. However, in some birds infected with C. galli, no clinical signs were observed [18]. Cryptosporidium meleagridis, originally described in birds, is the only Cryptosporidium species reported in both natural and experimental infections in avian and mammalian species, as well as humans [19].
Cryptosporidium avium, previously known as Cryptosporidium avian genotype V, was recognized as a valid species in 2016 [5]. The morphology, biology and host specificity have been studied. As the classification of species within the genus Cryptosporidium is constantly updated, more data was needed to support the establishment of a new separate species of the genus Cryptosporidium.
In the present study, we have revisited the infectivity and pathogenicity of C. avium using more animal species. Additionally, a new parasitic site has been discovered.

Source of oocysts
Cryptosporidium oocysts were obtained from the feces of naturally infected cockatiels (Nymphicus holandicus) in a pet market in the Province of Henan, China. Oocysts from cockatiels were pooled and used to infect 30 one-day-old Roman chickens. Oocysts from the 30 one-day-old Roman chickens were then used to infect other animals, after concentration using a water ether technique [20] and purification with discontinuous sucrose density centrifugation [21]. Oocysts were counted with a Neubauer hemocytometer. A combination of streptomycin and penicillin was added and this oocyst suspension was kept at 4°C.
Morphometry analyses of C. avium oocysts were performed using digital analysis of images (Motic Images Plus 2.0 software). A 20 μl aliquot containing 10 5 purified oocysts was examined for each measurement. The length and width of oocysts (n = 100) were measured under bright-field microscopy at 1000× magnification; these values were used to calculate the length-to-width ratio of each oocyst.

DNA extraction and molecular analyses
Genomic DNA was extracted using an E.Z.N.A.® Stool DNA Kit (Omega Bio-tek, Norcross, GA, USA). Extracted DNA was stored at -20°C until used in nestedpolymerase chain reaction (PCR) assays. Nested-PCR protocols were used to amplify partial sequences of the Cryptosporidium small-subunit rRNA gene (SSU), actin gene, HSP70 gene and Cryptosporidium oocyst wall protein gene (COWP), according to previous studies [22][23][24][25]. Negative (molecular grade water) and positive controls (DNA from C. baileyi) were included in each PCR amplification. Cryptosporidium species were also determined by PCR restriction fragment length polymorphism (RFLP) analysis, using SspI and VspI [8,26]. The PCR products were detected by agarose gel (1.5%) electrophoresis, purified with GenElute™ Gel Extraction Kit (Sigma Aldrich, St. Louis, MO, USA) and sequenced in both directions with secondary primers using a Big Dye Terminator v.3.1 cycle sequencing kit and ABI Prism 3130 genetic analyzer (Applied Biosystems, Carlsbad, CA, USA). The sequences were assembled using the ChromasPro 2.64 (http:// www.technelysium.com.au), and genotyped with a multiple-sequence alignment analysis together with reference sequences retrieved from the GenBank database using ClustalX 2.1 (http://www.clustal.org/).

Experimental design Animals
Thirty three-day-old and six 40-day-old Roman chickens, 30 ten-day-old quail (Coturnix coturnix japonica), 15 three-day-old Cherry Valley ducks, eight four-weekold BALB/c mice, eight four-week-old Mongolian gerbils (Meriones unguiculatus), and six 20-day-old budgerigars (Melopsittacus undulatus) were used for experimental infection studies. In addition, the same number of animals from each host species/strain was used as a negative control. All animals used in this study were obtained from the Henan Experimental Animal Center.
Each animal was confirmed to be free of C. avium infection by microscopic examination of feces. Animals were randomly divided into control and test groups, housed individually in plastic cages or bird cages under pathogen-free conditions, and received sterilized food and water. Each animal in the test group was inoculated orally by stomach tube with a dose of 1 × 10 6 oocysts suspended in 500 μl of distilled water. Each animal used as negative control was inoculated with equal doses of distilled water. One C. avium positive animal from the three-day-old chicken group was euthanized 15 days post-infection (dpi). Tissue samples were processed for histology and scanning electron microscopy. The light and electron microscopic examination were performed according to a previous study with a slight modification [27].

Clinical status
Each animal in each group was examined daily for the appearance of clinical signs. Rectal temperature, breath, appetite and presence of any abnormal behaviors were observed.

Oocyst excretion
Fecal samples were obtained daily during the experiment, starting from the second dpi, to determine the prepatent period. The experiments were terminated at 30 dpi. The number of oocysts in each animal of all experimentally infected groups was counted by hemocytometer slide under bright-field microscopy at 400× magnification. Dynamics of oocyst shedding were determined as a number of oocysts per gram (OPG). The OPG was estimated on the basis of the number of oocysts counted [28].

Histological examination
After a complete examination of all gastrointestinal organs at necropsy, organs and tissues collected from liver, trachea, stomach, duodenum, jejunum, ileum, cecum, colon and bursa fabricii (BF) were fixed in 10% buffered formalin for 24 h, dehydrated in absolute ethanol, cleared in xylene, and embedded in paraffin. Each tissue section was cut at a thickness of 4 μm, stained with hematoxylin and eosin (H&E), and observed microscopically at 1000× magnification by light microscopy.

Scanning electron microscopy
To further observe and confirm C. avium colonization in chickens, tissue samples from the BF were selected for scanning electron microscopy (SEM) observation according to the results of histological observation. Samples were fixed in 2.5% glutaraldehyde for one week at 4°C, and then washed with 0.1 mol/l phosphoric acid buffer (pH = 7.4) three times for 10 min each. The dehydration procedure followed conventional methods in a graded ethanol series of 30%, 50%, 70%, 90% and 100%, and two more changes of 100%, each for 5 min followed by 50% isoamyl acetate solution (v/v, isoamyl acetate: ethanol = 1:1) and 100% isoamyl acetate solution for 10 min, respectively. After specimens were critically point-dried using CO 2 and coated with gold, observations were made using an S-3400 SEM (Hitachi, Tokyo, Japan).

Clinical status
The appetites and attitudes of all animals in both experimental and control groups were normal during the experiment. All animals remained free of clinical signs at any point. No remarkable changes were observed in macroscopic observations. In addition, no animals died during the experiment.

Oocyst shedding
No Cryptosporidium oocysts were detected in the feces of experimentally inoculated BALB/c mice, Mongolian gerbils, quail or budgerigars within 30 dpi. However, fecal examination of chickens and ducks revealed fully sporulated C. avium oocysts. Oocysts were first detected in the feces of 3-day-old chickens at 8 dpi, peaking twice at 11 and 14 dpi. The infection intensity ranged from 2 × 10 3 to 8 × 10 4 OPG with maximum shedding at 14 dpi. In 40-day-old chickens, oocysts were first detected at 9 dpi. The infection intensity ranged from 1 × 10 3 to 3.5 × 10 4 OPG with maximum shedding at 11 dpi. Likewise, oocysts of C. avium were microscopically detected at 9 dpi in ducks, peaking twice at 12 and 17 dpi. The infection intensity ranged from 1 × 10 3 to 8 × 10 4 OPG with maximum shedding at 17 dpi. The patterns of oocyst shedding in chickens and ducks are presented in Fig. 2.

Molecular characterization
All isolates of C. avium (from naturally infected cockatiels, one-day-old Roman chickens, three-day-old Roman chickens, three-day-old ducks and 40-day-old Roman chickens) were positive for the SSU rRNA gene, actin gene and HSP70 gene by PCR, but the COWP gene was not present. The SSU rRNA gene, actin gene and HSP70 gene nucleotide sequence obtained in this study shared 100% identity with Cryptosporidium avian genotype V obtained from AY271721, AB471661 and AB538401, respectively.
The PCR-RFLP analysis indicated that the SSU rRNA gene of C. avium was cut into two pieces by SspI endonuclease, 497 bp and 253 bp, respectively (see Additional file 1: Figure S1a). By using the VspI endonuclease, the SSU rRNA gene of C. avium was cut into three pieces, 621 bp, 115 bp and 104 bp, respectively (see Additional file 1: Figure S1b). The sequences have been deposited in the GenBank database under accession numbers JQ246415 (SSU rRNA gene), JQ320301 (actin gene) and JQ798893 (HSP70 gene).

Histological and ultrastructural observation
The results of tissue section showed that Cryptosporidium infection in chickens (at 15 dpi) was only found in the epithelial cells of the BF (Fig. 3a, b). A large number of developmental stages of C. avium had adhered to the surface of the BF. This phenomenon was more pronounced in the longitudinal plica of the BF, when histologically observed by scanning electron microscopy. The BF was almost completely covered with C. avium at different endogenous stages (Fig. 3c-f). However, no Cryptosporidium developmental stages or pathological changes were observed in other host organs, including the liver, trachea, stomach or intestines. No pathological changes were observed.

Discussion
Cryptosporidium species are important zoonotic protozoans that infect a wide range of hosts [29]. More than 37 species of Cryptosporidium have been formally described and are considered valid [30]. Although several Cryptosporidium genotypes/isolates have been reported in birds, only four species are considered avian-adapted because of a lack of the biological and morphological data necessary for species designation [31,32]. Here, we revisit the infectivity and pathogenicity of C. avium and provide new information on the biology of this species.
In the present study, we have shown that the prepatent period of C. avium was eight and nine days in hens and ducks, respectively, similar to a previous study [5]. In chickens infected with C. baileyi, the prepatent period was three dpi [3]. In two nine-day-old chickens, oocysts were excreted for six consecutive days, beginning 25 days after feeding on C. galli oocysts [4]. The variability in the prepatent and patent period may depend on the species of Cryptosporidium and the status of experimental animals.
Differences in pathogenicity between Cryptosporidium species and genotypes have been reported in birds [38][39][40]. Cryptosporidium baileyi and avian genotype II are generally regarded as etiological agents for infections in the ocular conjunctiva, respiratory tract, BF, rectum and cloaca [41]. In domestic chickens and turkeys, infection with C. meleagridis causes subclinical infection or Fig. 1 Oocysts in fecal smears showed typical Cryptosporidium characteristics when stained with Modified Ziehl-Neelsen stain (a, ×1000). Oocyst morphology was also observed by differential interference contrast (DIC) microscope (b, ×1000) clinical signs related to intestinal infection [42,43]. Infection with C. galli, C. muris, and avian genotype III are characterized by chronic gastric disease, with clinical signs that include vomiting, weight loss, and macroscopic and microscopic lesions in the proventriculus [44,45]. In birds infected with C. avium, oocysts were detected in the kidney, ureter and cloaca in natural infections, and the ileum and cecum following experimental infection [5,37]. In this study, however, developmental stages of C. avium were mainly observed in longitudinal plica of the BF, with high numbers in histological sections and SEM study. This finding indicates that the BF may be the main parasitic site of C. avium infection.

Conclusions
We have revisited the infectivity and pathogenicity of C. avium in several species of animals. Compared to previous studies, oocysts of C. avium were mainly detected in the BF of three-day-old hens at 15 dpi. This reveals that the BF may be the main site of C. avium infection. All findings in the present study provide new information on the biology of C. avium.

Additional file
Additional file 1: Figure S1