Molecular epidemiology and cross-species transmission risk of Enterocytozoon bieneusi between humans and livestock: Evidence from Lishui, China

Abstract

Enterocytozoon bieneusi is a zoonotic parasite with a broad host range and public health significance. In China, livestock production is predominantly small-scale, with cattle and sheep commonly maintained under extensive or semi-intensive husbandry systems that lack adequate biosecurity measures. Lishui, Zhejiang Province, typifies this model, where intensive and non-intensive farming systems coexist, and where abundant rainfall and dense water networks facilitate pathogen transmission. A total of 588 fecal samples were collected from cattle (n = 175), sheep (n = 228), and humans (n = 185) across nine counties in Lishui. Nested PCR targeting the ITS region was used for detection and genotyping, followed by phylogenetic and haplotype network analyses. The overall infection rates were 32.9 % in sheep, 4.5 % in cattle, and 1.6 % in humans, with all human cases occurring in occupationally exposed farm workers. Significantly higher infection rates were observed in intensively managed herds and in young animals under one year of age (P < 0.05). Five genotypes were identified in sheep, among which BEB6 was predominant (80.0 %), while cattle harbored genotypes BEB8 and J. Human isolates comprised genotypes BEB6, J, and I. Phylogenetic analyses placed all identified genotypes within Group 2, and haplotype network reconstruction revealed 10 haplotypes, some of which were shared between human and livestock samples from the same farms. These findings highlight cross-species transmission risks under current farming practices and underscore the necessity for One Health-based surveillance and control strategies.

Highlights

• •Lishui's mixed farming and heavy rainfall elevate zoonotic risk.
• •Infection rates of E. bieneusi: sheep 32.9 %, cattle 4.5 %, humans 1.6 %.
• •All human cases were farm workers; genotypes BEB6, J, and I were detected.
• •Findings show cross-species transmission risk and call for One Health surveillance.

1. Introduction

Microsporidia are obligate intracellular eukaryotic parasites with extensive phylogenetic diversity and a broad host range, infecting both vertebrates and invertebrates (Stentiford et al., 2019). To date, nearly 1500 species across over 200 genera have been reported, among which Enterocytozoon bieneusi is the most commonly identified and clinically important species in humans (Li et al., 2019). Infections are often associated with diarrhea and malnutrition, with particularly severe outcomes in immunocompromised populations, including individuals living with HIV/AIDS and transplant recipients (Li and Xiao, 2021; Weber et al., 1999). Increasing evidence also indicates a considerable burden in immunocompetent individuals, especially children and rural residents in endemic regions (Li and Xiao, 2021; Mena et al., 2024; Mori et al., 2013).

The transmission pathways of E. bieneusi are complex, including direct person-to-person spread as well as fecal-oral transmission through contaminated water, food, or contact with infected animals (Pepper et al., 2006). A wide range of animals, particularly livestock, companion animals, and rodents, serve as reservoirs and contribute to environmental contamination (Pei et al., 2025). In some regions of China, where livestock farming is predominantly small-scale and biosecurity is limited, high infection rates have been documented in cattle, sheep, and goats (Fan et al., 2025; Hwang et al., 2020; Wegayehu et al., 2020), raising significant concerns for zoonotic spillover. Genotypes D, EbpC, and Type IV are frequently identified in both humans and animals, suggesting host-sharing patterns that imply potential for zoonotic transmission (Li and Xiao, 2021). Despite increasing reports of zoonotic transmission, there is still a lack of systematic molecular surveillance integrating human, livestock, and environmental data under a unified One Health framework, which limits the identification of dominant reservoirs, mapping of transmission pathways, and development of early-warning strategies.

Lishui City in southeastern China represents a typical mixed-scale livestock production region where intensive and extensive systems coexist within the same agricultural landscape. Weak biosecurity measures, poor fecal management, and high rainfall with dense river networks create favorable conditions for pathogen persistence and waterborne spread. These features highlight an urgent need to clarify zoonotic risks under such agro-ecological settings. In this study, we adopted a One Health framework to investigate the prevalence, genotype distribution, and cross-species transmission potential of E. bieneusi among livestock and humans across nine counties in Lishui. By situating the analysis within a region that reflects the diversity of Chinese smallholder farming systems, this work provides critical evidence for integrated surveillance and targeted interventions, thereby contributing to the broader One Health agenda for microsporidiosis control in rural environments.

2. Materials and methods

2.1. Sample collection

In November 2024, a total of 588 fecal samples were collected from 10 intensive and 13 small-scale non-intensive livestock farms across nine counties in Lishui City, Zhejiang Province (Suichang, Qingtian, Liandu, Jingning, Yunhe, Songyang, Longquan, Qingyuan, and Jinyun) (Fig. 1). The samples included 185 human samples (95 from farm workers and 90 from nearby residents), 175 from Chinese yellow cattle, and 228 from sheep. Among these, 44 cattle and 185 sheep samples were obtained from intensive livestock farm, while 131 cattle and 43 sheep samples were collected from non-intensive livestock farm (Table S1).

Fig. 1.

Geographic map of the sampling locations in Lishui, Zhejiang Province.

A stratified random sampling strategy was applied in the selection of farms, livestock, and surrounding human residents, ensuring representative coverage across different farming systems and geographical locations. All available farm workers at each site were included in the study, while nearby residents were randomly selected from households located within a 2 km radius of the sampled farms. All specimens were preserved in 2.5 % potassium dichromate solution and stored at 4 °C until further analysis.

2.2. DNA extraction and PCR amplification

The collected fecal samples were suspended in sterile distilled water and centrifuged at 1500 × g for 10 min at room temperature. The resulting fecal pellets were washed four times to remove residual potassium dichromate. Approximately 500 mg of each sample was used for DNA extraction using the Fast DNA Spin Kit (MP Biomedicals, Irvine, CA), following the manufacturer's protocol (Mo et al., 2024). All extractions were performed in a biosafety cabinet, and the extracted DNA was stored at −20 °C until further use.

Molecular identification of E. bieneusi was carried out using a nested PCR approach as described previously. Primer sequences, amplicon length, and annealing temperatures are listed in Table 1 (Xin et al., 2024). The target amplicon (392 bp) spans the terminal portion of the 18S rRNA gene, the complete internal transcribed spacer (ITS), and part of the 5.8S rRNA gene (Sulaiman Irshad et al., 2003). Each PCR run included a positive control (previously confirmed positive sample) and a negative control (sterile ultrapure water). The first- and second-round primers used were ITSF1/ITSR1 and ITSF2/ITSR2, respectively. PCR conditions were as follows: initial denaturation at 95 °C for 5 min, followed by 35 cycles of 95 °C for 30 s, 55 °C for 30 s, and 72 °C for 30 s, with a final extension at 72 °C for 7 min. Second-round PCR products were examined by agarose gel electrophoresis and visualized using GelRed (Biotium Inc., Hayward, CA) staining. Purified PCR products were sequenced bidirectionally by BGI Genomics (Beijing, China). Sequence quality was assessed using Chromas software. A standardized 243 bp fragment of the ITS region was trimmed and used for downstream alignment and genotype identification via NCBI BLAST (https://blast.ncbi.nlm.nih.gov/Blast.cgi), following the established nomenclature system. The sequences obtained in this study were deposited in GenBank under the accession numbers PV767494-PV767506.

Table 1.

The oligonucleotide primers used in this study.

Gene	Identification code	Sequences (5′-3′)	Annealing temperature (°C)	Length (bp)
ITS	F1	GAT GGT CAT AGG GAT GAA GAG CTT	55	392
	R1	TAT GCT TAA GTC CAG GGA G
	F2	AGG GAT GAA GAG CTT CGG CTC TG	55
	R2	AGT GAT CCT GTA TTA GGG ATA TT

2.3. Haplotype network and phylogenetic analysis

To determine the genetic structure of E. bieneusi in the study area, representative genotype sequences from various phylogenetic groups previously reported in GenBank were selected alongside sequences generated in this study. Multiple sequence alignment was performed using MEGA 11 software (version 11.0.13), and all sequences were trimmed to the 243 bp ITS gene region. A phylogenetic tree was constructed using the neighbor-joining method, with 1000 bootstrap replicates used to assess the robustness of the tree topology (Thompson et al., 1994). Genotype PtEbIX was used as the outgroup to root the tree.

To further evaluate the potential for transmission between humans and livestock, haplotype network analysis was conducted based on the 243 bp ITS gene region. Haplotypes were identified using DnaSP version 27, and a TCS network was constructed with a 95 % connection limit (Clement et al., 2000). The resulting haplotype network was visualized using PopART software (version 1.7), with the frequency of each haplotype annotated accordingly (Leigh et al., 2015).

2.4. Statistical analysis

Chi-square tests were applied to assess differences in E. bieneusi infection rates across groups, and to calculate p-values (P), odds ratios (OR), and 95 % confidence intervals (CI). P < 0.05 was considered statistically significant. All statistical analyses were performed using GraphPad Prism software (version 8.02; GraphPad Software Inc.).

3. Results

3.1. Infection of E. bieneusi in humans and livestock

Across nine counties in Lishui, E. bieneusi was identified in both human and livestock populations, and the detailed prevalence patterns are presented below.

3.1.1. Humans

E. bieneusi was detected in 3 of 185 individuals (1.6 %). All positive cases were from breeders working on intensive livestock farms, while no infections were identified in non-exposed residents. The infected individuals were located in Liandu District, Longquan County, and Jingning County (Table 2).

Table 2.

Occurrence of E. bieneusi in human, cattle and sheep in Lishui, Zhejiang Province.

City	Host	No. specimens	No. positive for E. bieneusi (%)	E. bieneusi genotype (no.)
Suichang	human	20	0 (0)
	cattle	19	0 (0)
	sheep	21	0 (0)
Qingtian	human	20	0 (0)
	sheep	54	11 (20.4)	BEB6 (5), CHG3 (6)
Liandu	human	26	1 (3.8)	BEB6 (1)
	cattle	8	0 (0)
	sheep	32	26 (81.3)	BEB6 (23), CM9 (1), COS-I (2)
Jingning	human	20	1 (5.0)	I (1)
	cattle	20	0 (0)
	sheep	20	0 (0)
Yunhe	human	20	0 (0)
	cattle	7	0 (0)
	sheep	33	11 (33.3)	BEB6 (6), CHG5 (4), CHG3 (1)
Songyang	human	20	0 (0)
	cattle	41	2 (4.9)	BEB8 (2)
Longquan	human	15	1 (6.7)	J (1)
	cattle	50	6 (12.0)	J (6)
Qingyuan	human	24	0 (0)
	cattle	15	0 (0)
	sheep	43	11 (25.6)	BEB6 (11)
Jinyun	human	20	0 (0)
	cattle	15	0 (0)
	sheep	25	16 (64.0)	BEB6 (15), COS-I (1)
Total	human	185	3 (1.6)	BEB6 (1), J (1), I (1)
	cattle	175	8 (4.6)	BEB8 (2), J (6),
	sheep	228	75 (32.9)	BEB6 (60), CM9 (1), CHG3 (7), CHG5 (4), COS-I (3)
		588	86 (14.6)	BEB6 (61), CM9 (1), CHG3 (7), CHG5 (4), COS-I (3), BEB8 (2), J (7), I (1)

3.1.2. Livestock

The overall infection in livestock was 20.6 % (83/403), with a significantly higher rate in sheep (32.9 %, 75/228) than in cattle (4.6 %, 8/175) (OR = 10.23, 95 % CI: 4.78–21.36, P < 0.05). Cattle infections were detected only in Songyang (4.9 %, 2/41) and Longquan (12.0 %, 6/50), with no significant difference between these counties (P > 0.05). Sheep infections were detected in five of the seven sampled counties, with the highest infection in Liandu (81.3 %) and Jinyun (64.0 %), followed by Yunhe (33.3 %), Qingyuan (25.6 %), and Qingtian (20.4 %) (Table 2). Rates in Liandu and Jinyun were significantly higher than those in other regions (P < 0.05).

3.2. Genotype distribution of E. bieneusi in humans and livestock

Across human and livestock populations in Lishui, multiple E. bieneusi genotypes were detected, and their distribution patterns are summarized below.

3.2.1. Humans

Three E. bieneusi genotypes were identified among the positive human cases. Genotype BEB6 was detected in Liandu, genotype J in Longquan, and genotype I in Jingning. Genotypes BEB6 and J were also detected in livestock from the same farms (Table 2).

3.2.2. Livestock

Across all livestock samples, seven genotypes were identified. Five genotypes were detected in sheep, namely BEB6, CM9, CHG3, CHG5, and COS-I. Among these, BEB6 was predominant, accounting for 80.0 % (60/75) of positive sheep samples, and was distributed across multiple counties. Notable geographic variation was observed: Liandu yielded BEB6, CM9, and COS-I; Jinyun harbored BEB6 and COS-I; Yunhe contained BEB6, CHG3, and CHG5; and both Qingyuan and Qingtian had BEB6 and CHG3. In cattle, only two genotypes were identified, with BEB8 occurring in Songyang and J in Longquan, with no overlap between the two counties (Table 2).

3.3. Risk factor analysis and human–livestock genotype concordance

A stratified analysis was performed to compare infection patterns of E. bieneusi across farming systems, host age groups, and levels of human-animal contact.

3.3.1. Farming systems

Animals raised under intensive conditions had a significantly higher infection of E. bieneusi infection rate (28.2 %, 66/229) than those managed in non-intensive systems (9.8 %, 17/174) (OR = 3.74, 95 % CI: 2.10–6.61, P < 0.05). All three infected people were all from intensive farms.

3.3.2. Age groups

Animals under 1 year of age exhibited a significantly higher infection rate compared with older individuals (OR = 5.91, 95 % CI: 3.512–9.921, P < 0.05). Species-specific analysis showed that young sheep (<1 year) had a markedly higher infection than adult sheep (OR = 7.6, 95 % CI: 4.044–14.27, P < 0.05), whereas no significant age-associated difference was observed in cattle (Table 3).

Table 3.

Results by age of animals tested for E. bieneusi infections.

Host	Age	No. tested	No. positive	OR (95 % CI)	P-value
Cattle	<1 year	51	2	0.8102 (0.1604–3.335)	P > 0.05
	>1 year	124	6
Sheep	<1 year	102	57	7.6 (4.044–14.27)	P < 0.05
	>1 year	126	18
Total	<1 year	153	59	5.91 (3.512–9.921)	P < 0.05
	>1 year	250	24

3.3.3. Genotype concordance

Genotypic comparison between infected farm workers and livestock under their care revealed complete genotype concordance on the same farms. In Liandu, the infected individual carried genotype BEB6, identical to that detected in 81.3 % (26/32) of sheep on the same farm, with BEB6 comprising 88.46 % of the positive samples. In Longquan, the farm worker carried genotype J, which was also detected in all six positive cattle (12.0 %, 6/50) from the same farm. Sequence analysis of the 243 bp ITS fragment confirmed 100 % identity between human and livestock isolates in both cases. In Jingning, the infected woker carried genotype I, which was not detected in the 10 animal samples collected from the same farm.

3.4. Phylogenetic and haploty pe network analyses reveal shared lineages between humans and livestock

Phylogenetic and haplotype network analyses showed the genetic relationships and distribution patterns of E. bieneusi isolates from humans and livestock in Lishui.

All isolates clustered within Group 2, a clade predominantly associated with ruminant hosts in previous epidemiological surveys (Fig. 2). Genotypes detected in human samples (BEB6, J, and I) were positioned on evolutionary branches closely related to the corresponding genotypes identified in local cattle and sheep.

Fig. 2.

Phylogenetic tree constructed using neighbor-joining (NJ) method of E. bieneusi based on ∼243 bp sequence of the internal transcribed spacer (ITS) region.

Haplotype network analysis of 86 positive ITS sequences identified 10 distinct haplotypes, with Hap_1 being the predominant variant, representing 63.9 % of all sequences. In sheep, Hap_1 was widely distributed and occurred in Qingtian, Liandu, Yunhe, Qingyuan, and Jinyun, indicating its broad dominance in ovine populations. Additional haplotypes in sheep included Hap_2 (Qingtian and Yunhe), Hap_3 and Hap_4 (Liandu), Hap_6 (Yunhe), Hap_10 (Qingyuan), and Hap_4 (Jinyun) (Fig. 3). In cattle, Hap_7 and Hap_8 were exclusively detected in Songyang County, whereas Longquan County exhibited Hap_9 as the sole detected haplotype (Fig. 3). Among human infections, the case from Liandu carried Hap_1, the case from Longquan carried Hap_9, and the case from Jingning carried Hap_5 (Fig. 3). Notably, Hap_1 and Hap_9 were present in both humans and livestock from the same counties.

Fig. 3.

Haplotype analysis results of E. bieneusi sequences detected in this study. Different colors represent different groups, including Qingtian sheep, Liandu sheep, Liandu human, Jingning human, Yunhe sheep, Songyang cattle, Longquan human Qingyuan sheep and Jinyun sheep, and the sizes of the circles represent the number of haplotypes.

4. Discussion

E. bieneusi is a globally distributed microsporidian parasite of zoonotic importance, increasingly noted for its broad host range and cross-species transmission potential (Li and Xiao, 2021). Although its impact in immunocompromised populations is well recognized, most studies have examined single host species, leaving the ecological and management factors that shape its transmission poorly defined (Mackenzie and Jeggo, 2019). By simultaneously analyzing human and livestock infections within the same agricultural landscape, this study provides new insights into how farming practices and local environmental conditions influence infection dynamics. Conducted in Lishui City, a region characterized by smallholder mixed-species farming, the findings highlight context-specific risks and underscore the value of One Health-based surveillance and control strategies.

Previous investigations have shown that E. bieneusi has wide infection rates in cattle and sheep, although reported infection levels vary considerably across regions. International surveys have documented rates exceeding 20 % in parts of the Middle East and South America, but generally below 5 % in North America and East Asia (Fayer et al., 2007; Fiuza et al., 2016; Hatam-Nahavandi et al., 2025). In China, sheep consistently show higher infection rates, often above 40 %, whereas cattle usually display lower and more variable rates depending on region and breed (Cheng et al., 2024; Duan et al., 2024; Fan et al., 2025; Qin et al., 2022; Shi et al., 2016). In this study, the prevalence in Lishui reached 32.9 % in sheep and 4.5 % in cattle, values lower than those reported in Jiangsu and Shanxi but clearly higher than previous data from Zhejiang, and comparable with some other provinces (Li et al., 2022; Xin et al., 2024). Notably, all cattle samples were from beef cattle, and the observed rate closely matched previous findings for this breed (Ma et al., 2015). These prevalence levels, observed under mixed-species and semi-intensive farming systems, indicate that parasite transmission can be sustained within livestock populations and that environmental contamination may play an important role in maintaining circulation. Genotype analysis provided additional evidence in support of this pattern. Seven genotypes were identified, including BEB6, CM9, CHG3, CHG5, and COS-I in sheep, BEB8 and J in cattle, and BEB6, J, and I in humans. All of these have been previously reported in other regions of China (Cheng et al., 2024; Fan et al., 2025; Qiu et al., 2019). Notably, genotypes I, J and BEB6 (Group 2) have previously been identified in human (Li and Xiao, 2021; Zhang et al., 2011), consistent with our detection in farm workers from Lishui, reflecting their broad distribution and occurrence in multiple host species. The presence of shared genotypes in both livestock and humans within the same farms suggests that certain E. bieneusi lineages are not host-restricted, increasing the potential for interspecies transmission under favorable ecological and management conditions.

This study confirms that farming practices and host age are key determinants of E. bieneusi transmission. Consistent with previous reports, younger animals (< 1 year) exhibited higher infection rates (Li et al., 2016; Ma et al., 2015), likely due to immunological immaturity and heightened susceptibility to environmental contamination (Cai et al., 2019; Taghipour et al., 2021). Livestock raised under intensive systems also showed significantly higher infection rates than those in free-range or non-intensive settings, suggesting that high stocking densities promote transmission via contaminated water, feed, or fomites (Guo et al., 2022; Hatam-Nahavandi et al., 2025; Li et al., 2022). Molecular evidence revealed genotype concordance between human and livestock isolates, with complete ITS sequence identity in Liandu and Longquan farms. In contrast, genotype I was detected in a farm worker from Jingning but not in the limited number of sampled animals from the same site, indicating that additional sampling is needed to clarify its source (Li and Xiao, 2021). Phylogenetic and haplotype analyses further showed clustering of human- and animal-derived isolates that were not always identical, which may indicate indirect transmission through shared environmental reservoirs. Together, these findings indicate that livestock management, sanitation, and human activity jointly shape the interface for cross-species spread, highlighting the need for surveillance strategies that integrate genotype distribution with local production conditions and strengthen basic control measures such as manure handling, water protection, and the use of protective equipment by farm workers.

A limitation of this study is that the modest sample size and geographic scope may not fully capture local transmission dynamics. In addition, the investigation was conducted at a single time point, precluding assessment of seasonal variation in E. bieneusi prevalence. Future work should incorporate environmental (water, soil) and wildlife sampling into multi-source molecular tracing and adopt longitudinal, multi-season sampling to better delineate pathogen flow across the human-animal-environment interface. Such integrated One Health approaches are crucial for establishing region-specific early warning and intervention strategies targeting both endemic persistence and emerging zoonotic risks.

5. Conclusions

This study systematically revealed the epidemiological characteristics and genotype distribution of E. bieneusi in humans, cattle, and sheep in Lishui City, Zhejiang Province. Sheep exhibited the highest infection rates, with distribution patterns influenced by host age and farming practices. Several genotypes were shared between humans and livestock, supporting the possibility of cross-species transmission within intensive farming systems. These findings highlight the need for targeted control measures integrating livestock management, occupational protection, and coordinated surveillance within the One Health framework.

Authors contributor

WH and WBY conceived the study and contributed the original idea. XLY, ZRM, QHM, JWQ, BX, WR, CML, JHZ, JXL, YY, and YWS performed the experiments. ZRM and QHM wrote the initial draft of the paper. XLY, WBY and WH contributed to the revision of the manuscript, and the final version was reviewed by WH. All authors approved the final manuscript.

CRediT authorship contribution statement

Xialiang Ye: Writing – original draft, Methodology, Conceptualization. Ziran Mo: Writing – review & editing, Writing – original draft, Visualization, Methodology, Conceptualization. Qinghan Meng: Writing – original draft, Methodology. Jingwei Quan: Methodology. Bin Xu: Methodology. Wei Ruan: Methodology. Jianhua Zhao: Methodology. Junxian Liu: Methodology. Cui mei Li: Methodology. Yang Yu: Methodology. Yuwei Shan: Methodology. Wenbin Yang: Writing – review & editing, Supervision, Funding acquisition, Conceptualization. Lei Xiu: Writing – original draft, Methodology, Conceptualization. Wei Hu: Writing – review & editing, Supervision, Project administration, Funding acquisition, Conceptualization.

Consent for publication

All participants consented to have their data published.

Ethics statement

The patients/participants provided their written informed consent to participate in this study, and animal sample collection was conducted with the consent of the farm owners.

Funding

This work was supported by Inner Mongolia Autonomous Region Science and Technology Leading Talent Team: Zoonotic Disease Prevention and Control Technology Innovation Team (2022SLJRC0023); Study on pathogen spectrum, temporal and spatial distribution and transmission features of the important emerging and re-emerging zoonosis in Inner Mongolia Autonomous Region (U22A20526); Key Technology Project of Inner Mongolia Science and Technology Department (2021GG0171); State Key Laboratory of Reproductive Regulation and Breeding of Grassland Livestock (2020ZD0008); Young Scientists Fund of the National Natural Science Foundation of China (32402911).

Declaration of competing interest

The authors declare that there is no conflict of interest regarding the publication of this paper.

Appendix A. Supplementary data

Supplementary material Table S1 Occurrence of E. bieneusi under different farming system