Introduction

Plague is a severe infectious disease caused by Yersinia pestis. As a statutory Category A infectious disease under the Law of the People’s Republic of China on the Prevention and Treatment of Infectious Diseases. The disease is characterized by acute onset, severe clinical course, strong transmissibility, and high case-fatality rate. Due to its high lethality, plague remains one of the priority targets for global public health security prevention and control [1]. As a natural focal disease, the long-term maintenance of plague in the natural environment relies on the synergistic interplay of four core components: the pathogen, vertebrate hosts, arthropod vectors, and suitable habitats [2]. Stable natural plague foci can only be established when all four components are fully satisfied. Among these key components, vertebrate hosts act as critical pathogen reservoirs and amplifiers, and their species composition, population density, and ecological niche directly determine the nature of plague foci and the risk of epidemic outbreaks.

Within the plague transmission cycle, hosts are generally categorized into primary and secondary hosts. Primary hosts represent the pivotal populations responsible for sustaining the enzootic circulation of Yersinia pestis in nature, and 14 primary host species have been documented in China’s natural plague foci [3]. Secondary hosts (also referred to as incidental hosts) do not contribute to the long-term enzootic maintenance of the pathogen, but can trigger acute epizootic outbreaks following infection and serve as important direct or indirect sources of human infection [3]. To date,72 secondary host species have been identified across China. In typical plague foci of the Qinghai-Tibet Plateau, primary hosts are predominantly rodent species such as marmots. Notably, Tibetan sheep (Ovis aries) have emerged as epidemiologically significant secondary hosts due to their high infection prevalence and documented association with human plague outbreaks, primarily via direct contact with infected or deceased sheep. Consequently, Tibetan sheep have become a core target for local plague surveillance and control interventions.

Surveillance data reveal that between 1958 and 2020,16 of 198 laboratory-confirmed human plague cases in Qinghai Province were attributed to direct or indirect transmission from Tibetan sheep [4], and fourteen strains of Yersinia pestis were isolated from Tibetan sheep [5]. Furthermore, human plague outbreaks linked to Tibetan sheep were documented in Longzi and Dangxiong, Xizang, in 1992 and 1994, respectively [6,7], underscoring the substantial role of Tibetan sheep in regional plague transmission dynamics. Gansu Province harbors two categories and three subtypes of natural plague foci: the Himalayan marmot plague focus in the alpine steppe of the Qilian Mountains-Altun Mountains, the Himalayan marmot plague focus in the alpine meadow steppe of southern Gansu, and the Alashan ground squirrel plague focus on the Loess Plateau of central Gansu. Among these, the Himalayan marmot plague focus in the alpine steppe of the Qilian Mountains-Altun Mountains remains epidemiologically active. Since 2000, a total of 10 human plague cases with 8 fatalities have been reported in this focus, corresponding to an 80% case fatality rate [8]. The majority of human infections were linked to direct contact with infected marmots or sheepdogs.

In the Qinghai-Tibet Plateau region, Tibetan sheep are susceptible to plague infection via two main routes: licking carcass bones of infected marmots, or being bitten by fleas that have migrated from marmot carcasses [4]. In recent years, sheep populations have expanded substantially in the Himalayan marmot plague foci of Gansu Province. The sheep inventory in Subei County increased from 337,860 head in 2020 to 414,304 head in 2022 [9]. According to field surveys, Tibetan sheep, as a confirmed secondary host of plague, have also increased in number through introduction. The role of sheep in plague ecology extends beyond contemporary transmission dynamics. Recent paleogenomic evidence has identified ancient Y. pestis lineages in prehistoric sheep—lineages otherwise found only in ancient humans—indicating that sheep served as important intermediate hosts in human plague transmission for at least 4,000 years [10]. This deep evolutionary context underscores the public health significance of monitoring plague in modern pastoral systems. However, despite the rapid expansion of Tibetan sheep populations in Gansu Province, the risk of plague infection and transmission in this species has not yet been systematically evaluated, leaving a significant research gap. This study addresses three specific questions: Do Tibetan sheep in Subei County exhibit serological evidence of Y. pestis exposure? Is the spatial distribution of seropositive herds associated with active marmot plague detection sites? What ectoparasite species are present on Tibetan sheep in this region, and do they include taxa documented on Himalayan marmots? Based on the empirical findings, targeted risk mitigation strategies will be proposed to provide a scientific evidence base for decision-making, thereby facilitating the prevention and control of plague outbreaks and safeguarding public health and social stability.

Research methods

Sample testing

Using the serum processed as described above, F1 antibody testing was performed in accordance with the operating instructions for the Yersinia pestis F1 Antibody Detection Kit and Appendix E of the Indirect Hemagglutination Assay (IHA) for Plague (WS279–2008) in the“Diagnostic Criteria for Plague” [11]. Both nasal swabs and ground samples of external parasites were subjected to molecular biological testing and bacterial culture. For molecular biological testing, nucleic acids were first extracted using a tissue genomic DNA extraction kit, followed by PCR amplification(Primers are shown in Table 1) and electrophoresis. The amplification system for each sample consisted of 15μl of 2 × PCR Mix,0.5μl of fraF, 0.5μl of fraR, 0.5μl of plaF, 0.5μl of plaR, 1μl of the sample to be tested, and 12μl of nucleic acid-free water; the amplification conditions were 5 min at 95°C for pre-denaturation, followed by 30 cycles of 1 min at 95°C,1 min at 55°C, and 1 min at 72°C, with a final extension of 5 min at 72°C. Bacterial cultures were performed using Hersh medium at 28°C for at least 24 hours, after which colony growth and morphological characteristics were observed. All detection reagents are listed in Table 2.

Download:

• PNG
  larger image
• TIFF
  original image

Table 1. Target gene design.

https://doi.org/10.1371/journal.pntd.0014650.t001

Download:

• PNG
  larger image
• TIFF
  original image

Table 2. Details of sample detection reagents.

https://doi.org/10.1371/journal.pntd.0014650.t002

Results

Sample collection

A total of 3,033 Tibetan sheep were sampled for this study, with 1,329 sampled in the spring and 1,704 in the autumn. The collected samples originated from Nan ning Guo le Village (926 sheep), Kui teng Guo le(185 sheep), and Diao er li ji(202 sheep) in Yan chi wan Township, Subei County; Jin gou(92 sheep) and Shi ban dun(32 sheep) in Shi bao cheng; Ying zui shan(16), Ha shi ha er(4), Yu er hong Pasture (1,102), and Yu er hong Village (474)—totaling 35 sheep herds from 9 villages (Fig 1, Table 3). Among these,4 herding sites were located at altitudes between 2,500 and 3,000 meters, while the remaining sites were situated between 3,000 and 4,000 meters. A total of 3,033 serum samples,1,090 nasal swab samples, and 36 Tibetan sheep (22 groups) were collected for examination of external parasites (Table 4).

Download:

• PNG
  larger image
• TIFF
  original image

Table 3. Number of Tibetan sheep samples collected in Subei Mongol Autonomous County, Gansu Province.

https://doi.org/10.1371/journal.pntd.0014650.t003

Download:

• PNG
  larger image
• TIFF
  original image

Table 4. Distribution of specimen types by village and herd.

https://doi.org/10.1371/journal.pntd.0014650.t004

Download:

• PNG
  larger image
• TIFF
  original image

Fig 1. Distribution of Tibetan sheep sampling sites and plague bacterium detection sites.

Red triangles indicate plague bacterium detection sites, green circles indicate spring sampling sites, and open circles indicate autumn sampling sites. The administrative division shapefile was obtained from Natural Earth (https://www.naturalearthdata.com/downloads/10m-cultural-vectors/), which is in the public domain. Elevation data were obtained from the Shuttle Radar Topography Mission (SRTM, https://earthexplorer.usgs.gov/), a public domain product of NASA. The map was produced in ArcGIS 10.8 (ESRI, USA).

https://doi.org/10.1371/journal.pntd.0014650.g001

Serological detection

Results of the indirect hemagglutination test for plague showed that 10 of the 3,033 Tibetan sheep serum samples tested positive for plague F1 antibody, with a positivity rate of 3.3‰ (10/3,033). Given the observed seroprevalence of 3.3‰, the sample size of 3,033 provides >99% power to detect a prevalence of ≥1‰ at α = 0.05 (two-tailed). However, the rarity of positive events (n = 10) precluded multivariable risk factor analysis for secondary objectives. Among these, the highest seroprevalence of plague antibodies was found in Tibetan sheep from Jin gou, at 10.8‰ (1/92); followed by Yu er hong Pasture, with a seroprevalence of 5.4‰ (6/1102); and Nan ning guo le, with a seroprevalence of 3.2‰ (3/926); no positive serum samples were detected in the remaining villages. Regarding the plague F1 antibody titers in the positive samples, titers ranged from 1:16–1:128, including 1 sample at 1:128, 1 at 1:64, 1 at 1:32, and 7 at 1:16. Overall, antibody titers were relatively low and distributed sporadically (Table 5). The 10 positive samples were collected from six different sheep flocks. Among them, five positive samples were detected in the flock of 387 Tibetan sheep belonging to Herder D, which had the highest positive rate of 13.0‰ (5/387); the positive rates for the flocks of the remaining five herders ranged from 2.5‰ to 9.5‰.

Download:

• PNG
  larger image
• TIFF
  original image

Table 5. Results of F1 antibody detection for plague in Tibetan sheep sera.

https://doi.org/10.1371/journal.pntd.0014650.t005

Ectoparasite detection

Combing examinations of Tibetan sheep for external parasites yielded a detection rate of 0.70% (21/3033). The 21 affected sheep were from five different flocks, with detection rates as follows: Flock B 0.70% (1/150), Flock C 2.30% (9/400), Flock G 64.0% (7/11), Flock H 2.24% (2/89), and Flock I 4.0% (2/50). The differences in the detection rates of ectoparasites among the different herds were statistically significant (Fisher’s Exact Test, P < 0.001).

A total of 36 ectoparasites were collected, with an overall parasite index of 0.012 (36/3,033). The breakdown by flock was: Flock B:2 parasites, index 0.013 (2/150); Flock C:17 parasites, index 0.043 (17/400); Flock G:13 parasites, index 1.182 (13/11); Flock H:2 parasites, index 0.022 (2/89); and Flock I:2 parasites, index 0.040 (2/50).

Seasonally, spring sampling (April–May) of 1,329 sheep identified 12 infested animals carrying 19 parasites, with a detection rate of 0.90% (12/1,329) and a parasite index of 0.014. Autumn sampling (October–November) of 1,704 sheep identified 9 infested animals carrying 17 parasites, with a detection rate of 0.53% (9/1,704) and a parasite index of 0.010. No statistically significant difference in ectoparasite detection rates was observed between seasons (χ² = 1.53, P > 0.050).

Ectoparasites were identified using standard taxonomic keys. The community composition comprising 6 hard ticks (Ixodidae), accounting for 16.70% (6/36), and 30 mites (Acari), accounting for 83.30% (30/36) (Table 6). Ectoparasite community structure was characterized by low diversity. The Shannon diversity index (H’) was 0.45 and the Simpson diversity index was 0.28, reflecting marked dominance of mites (83.3%) over ticks (16.7%).

Download:

• PNG
  larger image
• TIFF
  original image

Table 6. Taxonomic identification of ectoparasites on Tibetan sheep.

https://doi.org/10.1371/journal.pntd.0014650.t006

Discussion

As one of China’s three major coarse-wooled sheep breeds, the Tibetan sheep is an ancient indigenous breed widely distributed across high-altitude regions of the Qinghai-Tibet Plateau, including Tibet, Qinghai, Gansu, Yunnan, and Sichuan. Leveraging its geographical advantages as a highland interior and vast pasture resources, Qinghai Province has gradually developed into a core production area for Tibetan sheep. However, in recent years, Gansu Province has introduced Tibetan sheep to improve its sheep breed structure and increase production, and the scale of its herd is gradually expanding.

The grazing sites in this study spanned 2,500–4,000m, directly overlapping the core habitat of the Himalayan marmot (2,900–5,500m) [12], creating ecological conditions for cross-species Yersinia pestis transmission. Although a small number of marmot populations may descend along river valleys to riverbank grasslands at around 2,500 meters or to field ridges at around 2,800 meters, their core distribution area significantly overlaps spatially with the grazing areas of Tibetan sheep. This high degree of ecological niche overlap greatly increases the opportunities for contact between Tibetan sheep grazing areas and Himalayan marmots within the same geographical unit, creating natural conditions for the cross-species transmission of Yersinia pestis.

The 3.3‰ seroprevalence observed in this study aligns with the 0.9‰–10.7‰ (average 4.7‰) reported from multi-site surveillance in Qinghai Province [6], suggesting that plague exposure in Tibetan sheep is a stable component of the regional enzootic cycle rather than an isolated incident. Furthermore, by combining the patterns of antibody fluctuation in Tibetan sheep [6], the antibody half-life, and the 2-fold decay rule, it can be roughly estimated that the infection occurred between April 2024 and October 2025, which closely aligns with the marmot activity season (April–October annually). As an introduced species, Tibetan sheep have a relatively short history of grazing in the Jiuquan region of Gansu, yet the positivity rate has already reached 3.3‰. This indicates that once introduced into the epidemic focus, Tibetan sheep can be rapidly exposed to Yersinia pestis through direct contact, shared pastures, and vector bites. Although no human cases have been detected to date—likely due to factors such as the short duration of Tibetan sheep rearing, limited human contact, and the distance between marmot habitats and residential areas—this epidemic focus possesses a complete “marmot-Tibetan sheep-human” transmission chain. Human cases in this region have historically been linked to marmots and sheepdogs rather than sheep, likely due to closer human-dog contact and the potential for dogs to act as mechanical carriers or bridge hosts. Tibetan sheep may pose risk primarily through contact with infected carcasses or products (wool, meat) rather than direct zoonotic transmission [13]. Without proactive surveillance and an immunological barrier among livestock, activation of this chain would significantly increase the risk of sporadic or even clustered human outbreaks.

Vector organisms are essential intermediaries in the enzootic maintenance of Yersinia pestis, bridging primary rodent reservoirs and incidental hosts such as Tibetan sheep. In this survey, the overall ectoparasite burden was low and no fleas were recovered. The low yield likely reflects the cross-sectional design with spring and autumn sampling, which may have missed peak vector activity; additionally, some herders had recently applied acaricide dipping prior to our arrival. The ectoparasite index is influenced by various factors such as season, climate, and host density [14,15]; a single-season cross-sectional survey may not fully reflect vector dynamics. Future research should conduct continuous monitoring across multiple seasons and regions to obtain more reliable vector density data. Regarding the types of parasites detected, hard ticks (Ixodidae) and mites (Acari) are known to parasitize Himalayan marmots, and surveys in natural foci have shown that members of these taxa can carry Yersinia pestis [16]. The presence of these ectoparasites on Tibetan sheep is consistent with the hypothesis of vector-host switching from marmots; however, because specimens were identified only to family (Ixodidae) or subclass (Acari) level by morphological stereomicroscopy, species-level molecular confirmation would be required to fully support this conclusion. If such switching occurs, parasites detached from plague-killed marmots could transiently re-parasitize Tibetan sheep, and during outbreaks this transient host switching could facilitate interspecies pathogen transmission.

Molecular biology and bacterial culture tests of nasal swabs and parasite samples were all negative. Concurrently conducted positive and negative controls yielded corresponding results, indicating that the test results are reliable. Possible reasons for the negative sample results include: low pathogen load in the samples, below the limit of detection, resulting in negative results; studies have shown that different animal species exhibit significant differences in their response to Yersinia pestis infection, with some animals potentially presenting as subclinical infections or transient carriers [17]. If the sampling timing did not cover the peak period of bacterial shedding, it would be difficult to detect the pathogen. However, in conjunction with the results of the serological survey conducted concurrently with this study (3.3‰), it can be inferred that Tibetan sheep do indeed have a history of exposure to Yersinia pestis; however, pathogen evidence may not have been detected due to rapid clearance of the pathogen, inappropriate sampling timing, or limitations in the sensitivity of the detection method. This coexistence of serological positivity and negative pathogen detection is not uncommon in the surveillance of naturally occurring zoonotic diseases, suggesting that Tibetan sheep may primarily exhibit a “rapid clearance after infection” host pattern rather than a state of long-term carriage. Tibetan sheep may serve as “terminal hosts” or “sentinel animals” rather than “reservoir hosts.” Although they rapidly clear the pathogen after infection, they retain antibodies, which in turn confirms the activity of the epidemic focus; however, they themselves may not contribute to maintaining the focus. This distinction is crucial for the formulation of control strategies.

Conclusion

In summary, there is a risk of plague infection and transmission among Tibetan sheep and their vectors in the natural plague foci of the Himalayan marmot in Gansu Province. This study proposes the following targeted control measures: It is recommended that Tibetan sheep be included in the plague host surveillance system, with a focus on sampling and monitoring during plague outbreaks in marmots in endemic areas. Random sampling should be conducted on Tibetan sheep serum in pastoral areas, as well as at cattle and sheep trading markets and related biological materials. Serological and molecular biological methods should be used to detect plague F1 antibodies in the samples; Strengthen health education and behavioral interventions for high-risk groups (herders, sheep traders, and slaughterhouse workers) to enhance their awareness of and capacity for protection, thereby safeguarding their health. Concurrently, in-depth basic research on Tibetan sheep in plague-endemic areas should be conducted, exploring aspects such as infection mechanisms to provide a theoretical basis for optimizing plague prevention and control strategies.

Acknowledgments

This work was supported by the Natural Science Foundations of Gansu Province (24JRRA1157 to Dingsheng Wang and 25JRRA840 to Daqin Xu) and the Research Projects of Health Industry in Gansu Province (GSJKKY2025-15 to Daqin Xu and GSJKKY2025-06 to Aiwei He). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. No authors received a salary from any of the funders.