High prevalence and pathogenicity of Cryptosporidium serpentis in snakes in southern China

Abstract

In southern China, snakes have cultural and economic significance, serving both as traditional dietary resources and as increasingly popular pets. However, the prevalence and clinical impacts of Cryptosporidium spp. in snakes in southern China remain poorly understood. Between April 2018 and September 2020, we collected 357 fecal samples from wild snakes, farmed snakes, and pet snakes in Hunan and Guangdong, two provinces in southern China. Cryptosporidium spp. were identified and subtyped by sequence analyses of the small subunit rRNA (SSU rRNA) gene and the 60 kDa glycoprotein (gp60) gene, respectively. The intensity of oocyst shedding in Cryptosporidium-positive samples was evaluated using SSU rRNA-LC2 quantitative PCR. Histological examinations of gastric tissues from infected pet snakes were conducted to assess potential parasite-induced pathology. Overall, 93 of 357 (26.1%) samples were positive for Cryptosporidium spp., and the detection rates were 17.0%, 31.4%, and 46.2% in farmed snakes, pet snakes, and wild snakes, respectively. Five species of Cryptosporidium were identified, including C. serpentis (n = 77), C. tyzzeri (n = 6), C. varanii (n = 4), C. muris (n = 3), and C. parvum (n = 2). Only C. tyzzeri isolates were subtyped successfully and belonged to IXa subtype family. The highest average number of oocysts per gram (OPG) of feces was observed in C. serpentis samples (4.6 ± 1.7 logs), followed by C. varanii (3.5 ± 0.4 logs), C. tyzzeri (3.3 ± 1.0 logs), C. parvum (3.2 ± 0.4 logs), and C. muris (2.1 ± 1.7 logs). In pet snakes infected with C. serpentis, the gastric mucosal epithelial cells were heavily colonised by the parasites, resulting in significant damage to the villus structure. The results of this study indicate that C. serpentis is prevalent in snakes in southern China and has significant pathogenicity to snakes.

Graphical abstract

Highlights

• •93 out of 357 farmed snakes, pet snakes, and wild snakes in southern China were positive for Cryptosporidium spp.
• •Five species of Cryptosporidium were identified.
• •Cryptosporidium serpentis was the dominant species with high intensity of oocyst shedding.
• •Cryptosporidium serpentis was highly pathogenic to pet snakes.

1. Introduction

Cryptosporidium spp. are important zoonotic protozoan parasites that inhabit the gastrointestinal tract and respiratory epithelial cells of various vertebrate hosts (Checkley et al., 2015; Ryan et al., 2021). To date, 47 Cryptosporidium species and more than 120 Cryptosporidium genotypes have been recognized, and most of them are host-adapted (Ryan et al., 2021). Based on the preferred site of infection, Cryptosporidium species can be further classified into two major groups. The intestinal group includes most Cryptosporidium spp. that mainly parasitize the intestinal epithelium of hosts. The gastric group includes a few species that parasitize the gastric epithelium of their hosts, such as Cryptosporidium serpentis, Cryptosporidium muris, and Cryptosporidium andersoni (Fayer, 2010; Xiao, 2010).

Previous studies have identified several species of Cryptosporidium in snakes, including C. serpentis, C. muris, C. varanii, C. andersoni, C. parvum, and Cryptosporidium tyzzeri (Pedraza-Díaz et al., 2009; Rinaldi et al., 2012; Yimming et al., 2016). Cryptosporidium serpentis is the primary species identified in snakes and mainly causes subclinical symptoms, but occasionally causes a mid-body swelling (Paiva et al., 2013; Xiao et al., 2019). Additionally, snakes infected with C. serpentis may present with anorexia, regurgitation and weight loss (Paiva et al., 2013; Yimming et al., 2016). Cryptosporidium varanii parasitizes in the intestine of snakes and usually causes subclinical signs (Fayer, 2010; Xiao et al., 2019). Cryptosporidium tyzzeri and C. muris, which primarily infect rodents, are likely derived from rodents ingested by snakes; however, these species do not appear to establish active infections in snakes (Xiao et al., 2004a; Díaz et al., 2013). Therefore, species identification is important in assessing the potential damage caused by Cryptosporidium spp. in snakes.

Snakes have been widely bred in China and other Asian countries since 1990 for the high protein content of their meat and the medicinal value of their skin (Van Cao et al., 2014). Especially in the last two decades, snake farming in China and Southeast Asia has increased dramatically, with about 7000–9000 tons of snakes traded in China each year (Zhou and Jiang, 2004). In particular, annual snake meat consumption exceeds 3000 tons in Guangzhou, a city in southern China (Xiao et al., 2019). Additionally, snakes are becoming increasingly popular as pets due to their ease of care; this raises public health concerns regarding potential zoonotic transmission of pathogens between snakes and humans. Previously, several studies have been conducted to determine the occurrence of Cryptosporidium spp. in farmed snakes (Xiao et al., 2004b; da Silva et al., 2014; Karim et al., 2014; Xiao et al., 2019; Chai et al., 2021; Bogan et al., 2022; Ootawa et al., 2023). However, only a few studies have focused on pet snakes and wild snakes in China (Xiao et al., 2019; Zhang et al., 2020).

In the present study, we examined the occurrence of Cryptosporidium spp. in wild snakes, farmed snakes, and pet snakes in Hunan and Guangdong provinces of southern China, and identified Cryptosporidium species and subtypes present.

2. Materials and methods

2.1. Sample collection

Between April 2018 and September 2020, a total of 357 fecal samples were collected from four species of wild snakes, farmed snakes, and pet snakes in China, and stored in 2.5% potassium dichromate until DNA extraction. For farmed snakes, about 50 snakes were kept in a pen, and 5–8 fresh fecal samples were collected from different locations in the pen. For pet snakes, 1–2 snakes were kept in individual plastic boxes and thousands of snakes kept in a sealed room, and one fresh fecal sample was collected from each plastic box. For wild snakes, 26 samples were obtained through post-mortem intestinal collection. Of these samples, 159 were collected from farmed snakes (bred for meat or medical materials) on a farm in Hunan Province, including those from cobras (Naja nivea; n = 69), oriental rat snakes (Ptyas mucosus; n = 61), and stink rat snakes (Elaphe carinata; n = 29). The farmed snakes were fed with autoclaved formula food consisting of heat-processed poultry and livestock meat supplemented with vitamin premixes and mineral additives. The 172 samples from pet snakes were collected from corn snakes (Pantherophis guttatus) on a breeding farm. These snakes were fed on commercially sourced frozen-thawed mice. The pet snakes exhibited clinical signs upon sampling such as abdominal swelling. Twenty-six samples were collected from wild snakes inhabiting natural ecosystems in Guangzhou, Guangdong Province. The species and ages of these wild-caught snakes were not recorded because they were collected by local farmers without formal herpetological training. Farmed snakes and pet snakes investigated were divided into 3 convenient age groups: < 3 months-old, 3–6 months-old, and > 12 months-old. Fecal samples collected from snakes were all stored in 2.5% potassium dichromate (Sigma-Aldrich, Saint Louis, MS, USA).

2.2. DNA extraction and PCR analysis

Approximately 200 mg of each fecal sample was used to extract genomic DNA using the Fast DNA Spin Kit for Soil (MP Biomedical, Santa Ana, CA, USA) as previously described (Jiang et al., 2005). DNA was initially analyzed for Cryptosporidium spp. by PCR and sequence analysis of the small subunit rRNA (SSU rRNA) gene (Xiao et al., 1999). Cryptosporidium parvum and C. tyzzeri identified were further analyzed by the established 60 kDa glycoprotein (gp60)-based subtyping tool (Alves et al., 2003). However, gp60-based subtyping was not performed on the other three Cryptosporidium spp. identified (C. varanii, C. serpentis, and C. muris). This limitation was due to: (i) no gp60 subtyping tool has been developed for C. varanii; and (ii) the gp60 gene is absent in the published genomes of both C. serpentis and C. muris. The intensity of oocyst shedding in Cryptosporidium-positive samples was assessed using a SYBR Green-based SSU rRNA-LC2 qPCR performed on a LightCycler 480 II (Roche, Indianapolis, IN, USA) (Li et al., 2015). The number of oocysts per gram of feces (OPG) was calculated based on the Cq-values from the analyzed sample against a standard curve as described previously (Chen et al., 2019). The standard curve was generated by spiking Cryptosporidium-negative fecal samples with known concentrations (10¹–10⁶ oocysts/g) of C. parvum IOWA strain. The Cq-values from these spiked samples were plotted against log₁₀-transformed oocyst counts to establish a linear regression model, where Y represents the Cq-value and X represents oocyst concentration.

2.3. Sequence analysis

All positive secondary PCR products were sequenced bi-directionally in Sangon Biotech (Shanghai, China) to identify Cryptosporidium species/genotypes and subtypes. The raw nucleotide sequences were initially processed and assembled using ChromasPro 2.1.5.0 (http://technelysium.com.au/ChromasPro.html), followed by manual editing and alignment verification with BioEdit 7.1.3.0 (http://www.mbio.ncsu.edu/BioEdit/bioedit.html). For the identification of Cryptosporidium species and subtype, the edited sequences were subsequently aligned and compared against reference sequences of the SSU rRNA gene and the gp60 gene from GenBank using ClustalX 2.0.11 (http://clustal.org), respectively.

2.4. Histopathological examinations

Two pet snakes exhibiting abdominal swelling were euthanized, and their gastric tissues were collected for hematoxylin and eosin (HE) staining and scanning electron microscopy (SEM) examination. Tissues for HE staining were preserved in 10% buffered formalin overnight, dehydrated through a graded ethanol series, embedded in paraffin, sectioned, and stained with HE. Tissues for SEM were preserved in 2.5% glutaraldehyde, postfixed in 1% osmium tetroxide (OsO4), and examined under a Verios 460 (FEI, Oregon, USA) after critical point drying and gold sputtering.

2.5. Statistical analysis

Detection rates of Cryptosporidium spp. were compared in different age groups, snake species, and snake groups using the chi-square test implemented in SPSS v.20.0 (IBM Corp., New York, NY, USA). Differences were considered significant at P < 0.05.

3. Results

3.1. Prevalence of Cryptosporidium spp. in snakes

Of the 357 fecal samples collected from snakes, 93 (26.1%) were positive for Cryptosporidium spp. The detection rates of Cryptosporidium spp. in farmed snakes (17.0%; 27/159) were significantly lower than that in pet snakes (31.4%; 54/172; χ² = 9.3, df = 1, P = 0.0023) and wild snakes (46.2%; 12/26; χ² = 11.4, df = 1, P = 0.0007). In contrast, there was no significant difference in detection rates of Cryptosporidium spp. between pet snakes and wild snakes (χ² = 202, df = 1, P = 0.1368). Cryptosporidium spp. were detected in all three age groups of farmed and pet snakes. The detection rates of Cryptosporidium in snakes of < 3 months, 3–6 months and > 12 months were 24.0% (12/50), 26.9% (14/52), and 33.1% (55/166), respectively. There were no significant differences in the detection rates of Cryptosporidium spp. between different age groups (χ² = 0.8, df = 1, P = 0.3652; Table 1).

Table 1.

Occurrence of Cryptosporidium spp. and C. tyzzeri subtypes in snakes.

Location (province)	Source	Age	N	No. positive (%)	Cryptosporidium spp.					C. tyzzeri IXa
					C. serpentis	C. muris	C. varanii	C. parvum	C. tyzzeri
Hunan	Farmed snakes	< 3 months	21	9 (42.9)	9	0	0	0	0	0
		3–6 months	81	6 (7.4)	6	0	0	0	0	0
		> 12 months	57	12 (21.1)	12	0	0	0	0	0
		Subtotal	159	27 (17.0)	27	0	0	0	0	0
Guangdong	Pet snakes	< 3 months	29	3 (10.3)	1	0	0	2	0	0
		3–6 months	34	8 (23.5)	8	0	0	0	0	0
		> 12 months	109	43 (39.4)	31	3	4	0	5	2
		Subtotal	172	54 (31.4)	40	3	4	2	5	2
	Wild snakes	unknown	26	12 (46.2)	10	0	0	0	1	1
Total			357	93 (26.1)	77	3	4	2	6	3

Abbreviation: N, total number of samples.

3.2. Cryptosporidium spp. and C. tyzzeri subtypes in snakes

The species/genotypes of all 93 Cryptosporidium-positive samples were successfully identified by sequence analysis of the SSU rRNA gene. Five species of Cryptosporidium were identified, including C. serpentis (n = 77), C. tyzzeri (n = 6), C. varanii (n = 4), C. muris (n = 3), and C. parvum (n = 2). The SSU rRNA gene sequences of C. serpentis, C. tyzzeri, C. varanii, and C. parvum generated in the study were identical to the GenBank reference sequences EU553576, AF112571, PP088095, and LC012016, respectively. The three SSU rRNA gene sequences generated from C. muris-positive samples had a single nucleotide variation from the reference sequence DQ836341.

Of the six C. tyzzeri samples, three were successfully subtyped at the gp60 locus and belonged to subtype family IXa. The gp60 sequence of one C. tyzzeri sample was identical to the reference sequence GU951713, and the gp60 sequence of the other C. tyzzeri sample had a 12-nucleotide-deletion compared to the reference sequence GU951713. Subtyping at the gp60 locus of the two C. parvum samples failed. The failure to subtype the three C. tyzzeri samples and the two C. parvum samples is probably due to the low oocyst intensity in the samples.

3.3. Oocyst shedding intensity of Cryptosporidium spp

The intensity of oocyst shedding in snake positive for Cryptosporidium was assessed by SSU rRNA-LC2 qPCR. The OPG of C. serpentis (n = 77), C. tyzzeri (n = 6), C. varanii (n = 4), C. muris (n = 3), and C. parvum (n = 2) were 4.6 ± 1.6 logs, 3.3 ± 1.0 logs, 3.6 ± 0.4 logs, 2.1 ± 1.7 logs, and 3.2 ± 0.4 logs, respectively (Fig. 1A). In farmed snakes, the OPG of C. serpentis (n = 40) was 4.0 ± 1.7 logs. In pet snakes, the OPG of C. serpentis (n = 40), C. tyzzeri (n = 5), C. varanii (n = 4), C. muris (n = 3), and C. parvum (n = 2) were 4.8 ± 1.7 logs, 3.8 ± 0.3 logs, 3.6 ± 0.4 logs, 2.1 ± 1.7 logs, and 3.2 ± 0.4 logs, respectively. In wild snakes, the OPG of C. serpentis (n = 10) and C. tyzzeri (n = 1) were 4.3 ± 0.7 logs and 2.4 logs, respectively (Fig. 1B).

Fig. 1.

Oocyst shedding intensity (oocysts per gram of feces, OPG) of different species of Cryptosporidium in snakes recovered during this study (mean ± standard deviation). A The OPG of Cryptosporidium spp. in all snake groups. B The OPG of Cryptosporidium spp. in farmed snakes, wild snakes, and pet snakes.

3.4. Histopathological lesions

Histological examination of gastric tissues of C. serpentis infected pet snakes showed significant parasite burden and severe lesions. The architecture in the stomach of pet snakes appeared with granular cell atrophy, with the epithelial surface heavily colonized by Cryptosporidium (Fig. 2A and B). The scanning electron microscope images confirmed the heavy burden of Cryptosporidium oocysts in snake stomachs (Fig. 2C and D). These histological results are similar to previous studies in captive snakes (Brownstein et al., 1977).

Fig. 2.

Parasite load in the stomach of pet snakes infected with C. serpentis. A, B Parasite burden and histological changes revealed by hematoxylin-eosin staining (HE). C, D Parasite burden revealed by scanning electron microscopy (SEM).

4. Discussion

The results of the present study indicate that Cryptosporidium spp. are prevalent in snakes in the Chinese provinces of Hunan and Guangdong. Overall, 26.1% of the snakes investigated were positive for Cryptosporidium spp., with a significantly lower detection rate of Cryptosporidium in farmed snakes (17.0%) than in pet snakes (31.4%) and wild snakes (46.2%). The low detection rate in farmed snakes may be due to the autoclaved formula food. In addition, the detection rate in pet snakes (31.4%) in this study was higher than that observed in other studies with similar number of samples and employing the identical PCR protocol conducted in Beijing (4.4–6.2%) and Chengdu (16.7%) in China, and in Italy (4.4%) (Rinaldi et al., 2012; Chai et al., 2021; Zhang et al., 2020). The high detection rate of Cryptosporidium spp. in pet snakes may be due to the highly intensive farming model. In contrast, the detection rate of farmed snakes (17.0%) in Hunan was comparable to that observed in Thailand (17.6%) and Japan (26.0%) (Kuroki et al., 2008; Yimming et al., 2016), but higher than those reported in Dazhou (0%), Ziyang (0%), and Wuhan (10.1%) in China employing the identical PCR (Xiao et al., 2019; Chai et al., 2021). The different detection rates in farmed snakes are likely attributable to variations in the number of samples and samples sources. The detection rate in wild snakes (46.2%) in this study was higher than that observed in other studies conducted in Hubei (2.0%) (Xiao et al., 2019). The higher detection rates of Cryptosporidium spp. observed in wild animals in our study, compared to those reported in Hubei Province, likely reflect the predatory behavior and ecological exposure of the wild snakes. These wild snakes frequently prey on some Cryptosporidium hosts, such as lizards and snakes, which themselves have elevated infection risks due to continuous exposure to environmental oocysts in their natural habitats. In addition, the time of sampling may affect the detection rate in wild snakes, and samples collected three days after meals may actually have a higher detection rate than this study.

An obvious difference was observed in the distribution of Cryptosporidium species between pet snakes and farmed snakes. In this study, C. serpentis was the only Cryptosporidium species found in farmed snakes. This differs from the observations of most studies in the world, except for one study in Brazil (Ruggiero et al., 2011). In contrast, C. serpentis, C. varanii, C. parvum, C. muris, and C. tyzzeri were identified in pet snakes, and is similar to observations of previous studies (Díaz et al., 2013; Yimming et al., 2016). The different distribution of Cryptosporidium spp. between farmed snakes and pets may be due to the feeding sources. In this study, farmed snakes were fed on formula feed, but pets snakes were fed on quick-frozen mice. Snakes were fed on formula feed that conduce to reduce the spread of Cryptosporidium. The three non-snake specific Cryptosporidium species identified in pet snakes, C. parvum, C. muris and C. tyzzeri, are common in rodents (Zhang et al., 2022). Therefore, the identification of these three species in pet snakes may be due to ingestion of frozen-thawed mice infected by the species. Cryptosporidium parvum, C. muris, and C. tyzzeri do not appear to establish active infections in snakes. In contrast, only snake-specific C. serpentis was identified in farmed snakes fed on formula feed.

Cryptosporidium serpentis poses a major threat to snake farming. In this study, the detection rate of C. serpentis was very high in pet snakes, farmed snakes, and wild snakes. Furthermore, C. serpentis was the only species of Cryptosporidium in farmed snakes and was the dominant species in pet snakes and wild snakes, similar to previous studies (Xiao et al., 2004b; Rinaldi et al., 2012; Díaz et al., 2013). Additionally, a large number of C. serpentis were observed in the stomach and feces of positive snakes. These indicated a high infectivity of C. serpentis to snakes. More importantly, the large number of parasites colonizing the stomach caused serious damage to the gastric tissue, leading to obvious clinical signs in the snakes. Therefore, the high infectivity and pathogenicity of C. serpentis to snakes pose a great threat to snake growth.

Pet snakes may play a potential role in the zoonotic transmission of Cryptosporidium spp. In this study, four species of Cryptosporidium were identified in pet snakes. Except for C. serpentis, the other three species (C. parvum, C. muris, and C. tyzzeri) have all been reported in humans (Ryan et al., 2014). Cryptosporidium parvum is one of the two dominant Cryptosporidium species infecting humans and has caused numerous cryptosporidiosis outbreaks in humans (Feng et al., 2018). Cryptosporidium muris mainly infect rodents, but has been involved in numerous human cryptosporidiosis cases, especially in immunocompromised patients (Gharpure et al., 2019). Cryptosporidium tyzzeri is commonly identified in house mice but is occasionally identified in humans (Stensvold et al., 2024). Additionally, the isolates of C. tyzzeri identified in this study belong to the IXa subtype family, which was previously reported in human cases in Kuwait and the Czech Republic (Sulaiman et al., 2005; Rasková et al., 2013). Previous studies have shown that Cryptosporidium oocysts were not infectious when frozen at −20 °C for more than 24 h (Fayer and Nerad, 1996). Therefore, reducing the zoonotic risk of pet snakes can be achieved by feeding them mice that had been frozen at −20 °C for more than 24 h.

Our findings demonstrate that proper management protocols can significantly reduce Cryptosporidium spp. infection rates, as evidenced by the distinct patterns observed between farmed and pet snakes. Farmed snakes exhibited lower infection rates, with only the snake-adapted C. serpentis detected and a complete absence of non-snake-adapted Cryptosporidium spp. (C. parvum, C. tyzzeri, and C. muris). This likely results from their exclusive formula-based diet, which eliminates prey-associated transmission routes. In contrast, C. serpentis in addition to other non-snake-adapted Cryptosporidium spp. were detected in pet snakes, which is potentially due to consumption of improperly processed frozen-thawed rodents that may serve as transmission vectors. This observation has been reported by Fayer and Nerad (1996), who demonstrated that freezing at −20 °C for more than 24 h effectively inactivates Cryptosporidium oocysts. These findings collectively highlight the importance of comprehensive quarantine measures that incorporate both animal isolation protocols and strict food safety controls, including either formula feeding or proper freezing of prey items.

5. Conclusions

This study reported the prevalence of Cryptosporidium in snakes in Hunan and Guangdong, China. The results indicated that Cryptosporidium spp. were prevalent in snakes. Cryptosporidium serpentis was the dominant species in the snakes, and the heavy colonization by the parasite resulted in severe damage to the villus structure in the stomach. Therefore, C. serpentis is a major threat to snake farming. Further studies are needed to monitor the molecular epidemiology and pathogenesis of Cryptosporidium spp. infection in farmed snakes.

CRediT authorship contribution statement

Falei Li: Data curation, Investigation, Methodology, Writing – original draft, Writing – review & editing. Xinrui Wang: Investigation, Writing – review & editing. Lihua Xiao: Data curation, Writing – review & editing. Yaoyu Feng: Conceptualization, Data curation, Funding acquisition, Project administration, Writing – original draft, Writing – review & editing. Yaqiong Guo: Conceptualization, Data curation, Funding acquisition, Project administration, Writing – original draft, Writing – review & editing.

Ethical approval

Fecal samples were collected from farmed snakes and pet snakes with the permission of the animal owners. Sampled animals were handled according to the established procedures of the Chinese Laboratory Animal Administration Law of 2017. The research protocol was reviewed and approved by the Research Ethics Committee of the South China Agriculture University.

Data availability

The data supporting the conclusions of this article are included within the article. Representative newly generated nucleotide sequences were submitted to the GenBank database under the accession numbers PQ825952-PQ825956.

Funding

This study was supported by the National Natural Science Foundation of China (32273032), Guangdong Major Project of Basic and Applied Basic Research (2024B1515020116), 111 Project (D20008), Specific University Discipline Construction Project (2023B10564003), and Key Projects of Scientific Research Plan of Colleges and Universities of Anhui Province (2023AH050427).

Declaration of competing interests

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.