We collected adult ticks by environmental dragging or removed ticks from sheep (Ovis aries) at 6 locations from the Bayan-Khongor, Selenge, and Umnugovi provinces in Mongolia during April 2021–May 2022. We morphologically identified, e-vouchered, and processed the tick samples as described elsewhere (3). We used nucleic acids from individual ticks for DNA barcoding and generic virus screening (3–5), then used the results to randomly assign each to a single or pooled sample (pool), in which we combined 5 μL of individual tick nucleic acids according to species and locality into 2–12 ticks per pool (Appendix 1 Figure 1).

We performed nanopore-based metagenomic sequencing as described elsewhere (3). We performed similarity searches, de novo assembly, and read mapping on reads of >200 bp in Geneious Prime version 2022.2.1 (https://www.geneious.com). For mapping and pairwise comparisons, we used Rickettsia reference genomes (Appendix 2 Table 1).

In this study, we used a metagenomic sequencing–based approach to detect and characterize tickborne pathogens agnostically in comparable numbers of single-tick and pooled tick samples, representing 3 tick species endemic to Mongolia. We detected pathogenic bacteria, viruses, or parasites in 86.7% of samples with a predominance of bacterial pathogens (Table 1). The most frequently observed bacteria were SFG Rickettsia (67.6%) and related endosymbionts (45.5%). Bacteria of the genus Rickettsia, gram-negative obligate intracellular bacteria, account for most bacterial infections transmitted by ticks (8). SFG includes >30 distinct Rickettsia species associated mainly with symptoms of spotted fever. Previous reports using various detection approaches have documented several Rickettsia species associated with spotted fever in Mongolia and the neighboring Inner Mongolia Autonomous Region of China (2). We identified R. raoultii, R. slovaca, and R. canadensis in single-tick and pooled tick samples from Mongolia. Initially reported in Haemaphysalis leporispalustris ticks from Canada, R. canadensis was also detected in Haemaphysalis japonica ticks from far eastern Russia, Haemaphysalis flava ticks from South Korea, and Haemaphysalis longicornis and I. persulcatus ticks from China. Monophyletic clustering based on ompB and gltA genes, observed in this study and elsewhere, suggests that R. canadensis constitutes an independent group (9). Although its pathogenicity is unclear, serologic evidence of R. canadensis exposure in humans has been documented (10), making the pathogen a potential agent of tickborne infections in endemic regions, including Mongolia.

Other bacterial pathogens detected in ticks, including A. ovis, C. burnetii, and B. miyamotoi, have been documented previously in various regions in Mongolia (2). We further detected F. tularensis subsp. holarctica, a subspecies that causes human tularemia, and F. tularensis subsp. novicida in Hy. asiaticum ticks collected from different locations. In contrast to tularemia agents, F. tularensis subsp. novicida rarely causes human disease and, even then, mostly involving persons with coexisting medical conditions or immunosuppression (11). Cases of human tularemia have been reported in Mongolia; therefore, our findings documenting the presence of Francisella species bacteria associated with human infections helps define the epidemiology of this pathogen. Finally, we identified the intracellular mollicute S. ixodetis in pooled I. persulcatus tick samples in Mongolia. Although members of the genus Spiroplasma are vertically transmitted endosymbionts of Ixodid ticks, cases of acute febrile illness caused by S. ixodetis have been documented in both immunocompromised and immunocompetent adults with frequent tick exposure (12). In addition to those bacteria, we identified the Piroplasmorida apicomplexans parasites B. microti and T. equi, previously reported to circulate in various locations in Mongolia (2). Despite the presence of those and other potentially pathogenic species, no information on human infections is currently available.

Tickborne viral pathogens detected in the study include CCHFV, BJNV, and ALSV in 11.7% (8/68) of samples. We identified sequences of CCHFV in low abundance in individual Hy. asiaticum tick samples (Appendix 2 Table 2). Previous screenings carried out in various regions have described a low frequency of CCHFV in Hy. asiaticum ticks (6). Of note, relatively high CCHFV seroprevalence in humans has been documented, as well as detectable antibodies in various mammals that might serve as reservoirs, although no human cases have been documented to date (6). BJNV has recently been included as a species in Norwavirus genus (family Nairoviridae), members of which lack the medium genome segment encoding for the structural glycoproteins in other nairoviruses, such as CCHFV (6). BJNV was described in tick-associated human febrile disease of unknown etiology from the Inner Mongolia Autonomous Region of China, further displaying pathogenicity in cell lines and experimental infections (13). Virus-specific antibodies were detected with a prevalence of up to 54.6% in human convalescent serum, as well as in sheep and cattle from the region. Detection of virus nucleic acids were reported from several tick species in China (13). BJNV was the most prevalent tickborne virus in our study at a 5.8% detection rate; we found the virus exclusively in I. persulcatus tick pools (Table 1).

ALSV is another recently described virus with human health effects. It was originally described in in patients in the Inner Mongolia Autonomous Region and Heilongjiang Province of China with febrile diseases and a history of tick bites, followed by seroconversion (14). ALSV particles are enveloped, possessing positive-sense single-stranded RNA genomes in 4 segments and classified in the Jingmenvirus group of Flaviviridae because of nonstructural protein homologies. ALSV has been reported in various tick species collected in several countries in Eurasia and showed evidence of viral replication and exposure in sheep, cattle, and deer (7,15). We detected ALSV in 2 I. persulcatus tick pools and assembled the prototype virus genome (ALSV-Mongolia-b52). Maximum-likelihood analyses revealed grouping of ALSV-Mongolia-b52 with viruses reported from humans and ticks from China. Further analysis of the putative virus VP1a and VP1b surface glycoprotein sequences revealed amino acid markers associated with ASLV isolates from Asia hosted by I. persulcatus ticks (Table 2) (7). Those findings confirm the presence of ASLV in Mongolia and indicate I. persulcatus ticks are a probable vector. As with BJNV and many other tickborne infections, human ALSV infections are reported as tick bite–associated, nonspecific febrile illnesses. Diagnostic assays are imperative to determine the public health burden of those emergent human pathogens.

Among limitations, the study cohort was composed of 3 tick species that represent frequently documented tickborne pathogen vectors, including those species mostly observed in the northern part of Mongolia that is covered by boreal forest (I. persulcatus ticks), in the northern and central steppes (D. nuttalli ticks), and in the southern aimag areas (Hy. asiaticum ticks) (2). Nevertheless, we have described several other tick species within the genera Dermacentor, Haemaphysalis, Hyalomma, Ixodes, and Rhipicephalus, although some have been observed only rarely and include scarce information on public health effects (2). Our screening strategy included molecular barcoding and generic virus screening for individual ticks, whereas we pooled randomly selected samples of identical tick species and collection site. We performed microbial characterization of prescreened single-tick and pooled tick samples using metagenomic sequencing. Despite random assignment of ticks for individual or pooled screening, target genome quantity and infection rates of particular microorganisms might further hamper detection and might not correlate with actual prevalence. Therefore, we presented only pathogen detection rates and did not calculate the maximum-likelihood estimates or minimum infection rates, which are frequently generated based on data from target-specific assays. Nevertheless, incorporating morphology and barcoding for accurate species identification and generic testing as needed, combined with metagenomic sequencing for accurate pathogen characterization is an efficient strategy. Because of limited sample availability, we could not perform a systematic comparison of single-tick versus pooled tick testing or pathogen detection rates between collection sites. However, our findings demonstrate the utility of a metagenomic sequencing approach for detecting and characterizing both endemic and emerging pathogens, especially for locations with limited pathogen information.

In conclusion, we documented several bacterial and viral pathogens, some of which were initially described in ticks from Mongolia. Although the public health impact of those pathogens remains unclear, our findings demonstrate that nanopore sequencing can aid in detecting endemic and emerging tickborne pathogens.