Surveillance of tick-borne bacteria infection in ticks and forestry populations in Inner Mongolia, China

Ticks are one of the most important vectors that can transmit pathogens to animals and human beings. This study investigated the dominant tick-borne bacteria carried by ticks and tick-borne infections in forestry populations in Arxan, Inner Mongolia, China. Ticks were collected by flagging from May 2020 to May 2021, and blood samples were collected from individuals at high risk of acquiring tick-borne diseases from March 2022 to August 2023. The pooled DNA samples of ticks were analyzed to reveal the presence of tick-borne bacteria using high-throughput sequencing of the 16S rDNA V3–V4 region, and species-specific polymerase chain reaction (PCR) related to sequencing was performed to confirm the presence of pathogenic bacteria in individual ticks and human blood samples. All sera samples were examined for anti-SFGR using ELISA and anti-B. burgdorferi using IFA and WB. A total of 295 ticks (282 Ixodes persulcatus and 13 Dermacentor silvarum) and 245 human blood samples were collected. Rickettsia, Anaplasma, Borrelia miyamotoi, and Coxiella endosymbiont were identified in I. persulcatus by high-throughput sequencing, while Candidatus R. tarasevichiae (89.00%, 89/100), B. garinii (17.00%, 17/100), B. afzelii (7.00%, 7/100), and B. miyamotoi (7.00%, 7/100) were detected in I. persulcatus, as well the dual co-infection with Candidatus R. tarasevichiae and B. garinii were detected in 13.00% (13/100) of I. persulcatus. Of the 245 individuals, B. garinii (4.90%, 12/245), R. slovaca (0.82%, 2/245), and C. burnetii (0.41%, 1/245) were detected by PCR, and the sequences of the target genes of B. garinii detected in humans were identical to those detected in I. persulcatus. The seroprevalence of anti-SFGR and anti-B. burgdorferi was 5.71% and 13.47%, respectively. This study demonstrated that Candidatus R. tarasevichiae and B. garinii were the dominant tick-borne bacteria in I. persulcatus from Arxan, and that dual co-infection with Candidatus R. tarasevichiae and B. garinii was frequent. This is the first time that B. miyamotoi has been identified in ticks from Arxan and R. solvaca has been detected in humans from Inner Mongolia. More importantly, this study demonstrated the transmission of B. garinii from ticks to humans in Arxan, suggesting that long-term monitoring of tick-borne pathogens in ticks and humans is important for the prevention and control of tick-borne diseases.


Introduction
Ticks are obligate hematophagous ectoparasites of terrestrial vertebrates that can transmit various pathogens to humans and are considered to be the second most notable vectors worldwide after mosquitoes (1).By the end of 2018, 103 tick-borne pathogens were detected in China, of which 65 were newly identified in the past two decades (2).Notable bacteria such as Rickettsia, Borrelia, Coxiella, Anaplasma, and Ehrlichia (2-4), have been detected in a wide range of tick species that can cause spread of relevant zoonotic diseases.Annually expanding geographic ranges and increasing populations of ticks, partly driven by climate change, demographic growth, and increased overseas travel and trade, have become serious threats to public health (5,6).However, systematic and comprehensive investigations of tick-borne pathogens have not been conducted in tick habitats in China because of the limitations of investigation regions and research technology.
Inner Mongolia Autonomous Region, located in northern China, is well-known for its various landscapes and is one of the natural foci of various tick-borne diseases in China (7)(8)(9)(10).Arxan, on the border between China and Mongolia in the northeastern Inner Mongolia, features abundant wildlife resources and a developed livestock industry that provides a suitable environment for ticks to survive and reproduce.Trade activities at border crossings and tourism have partly increased the risk of tick-borne diseases.For instance, the average annual incidence of tick-borne encephalitis (TBE) in Arxan from 2006 to 2013 was 188.48/1000000, indicating that Arxan has the highest incidence of TBE in China (11).However, tick-borne bacteria in Arxan have not been studied in recent years.Therefore, investigating the status of bacterial infections in ticks and humans plays an important role in the prevention and control of tick-borne diseases in Arxan.
In this study, we performed high-throughput sequencing of the 16S rDNA V3-V4 region to detect tick-borne bacteria carried by ticks, and species-specific polymerase chain reaction (PCR) was performed to confirm the presence of pathogenic bacteria in ticks and participants at a high risk of tick-borne diseases in Arxan.Furthermore, antibodies against dominant tick-borne bacteria were detected using Enzyme-Linked Immunosorbent Assay (ELISA), Indirect Immunofluorescence Assay (IFA), and Western Blotting (WB) assays, providing complete and comprehensive monitoring of tick-borne bacteria and a theoretical basis for the prevention and control of tick-borne diseases.

Collection and identification of ticks
Free-living ticks were collected from ground vegetation by dragging white cloth in May 2020 and 2021 in Arxan, Inner Mongolia, China (Figure 1).All collected ticks were placed in breathable and moistened bottles and transported to the laboratory as soon as possible at room temperature (20-25°C).Tick species were identified based on morphological characterization and verified using sequences of mitochondrial 16S rDNA genes, as previously described (12).After identification, the ticks were stored at −80°C until use.

Collection of human blood samples
Patients with tick-borne diseases mostly work in the field, including as forestry workers (13-15).These workers had a high possibility of tick exposure and therefore, were defined as groups at high risk of acquiring tick-borne diseases.We simultaneously collected anticoagulated (EDTA blood) and non-anticoagulant blood samples from 245 participants (232 males and 13 females) working in the field from March 2022 to August 2023.The workplaces of these participants were Tianchi forest farm (8

DNA extraction
All ticks were washed with 75% ethanol and rinsed thrice with sterile water to remove environmental contaminants.Except for the 182 I. persulcatus samples used for the isolation and culture of B. burgdorferi, the remaining 100 I. persulcatus and 13 D. silvarum samples were used for DNA extraction.Individual tick samples were homogenized in 180 μL buffer ATL, then genomic DNA was extracted using the DNeasy Blood & Tissue Kit (QIAGEN, Germany) according to the manufacturer's instructions.Similarly, the genomic DNA of human blood samples was extracted from anticoagulated blood using the DNeasy Blood & Tissue Kit (QIAGEN, Germany).Each time DNA extraction was performed, an extraction control (water) was added.All DNA samples from both ticks and human blood were stored at −80°C until use.

High-throughput sequencing
Individual DNA samples of ticks were mixed in an equal volume (10 μL) to prepare pooled DNA samples (3 or 4 DNA samples for a pool).One hundred I. persulcatus DNA samples were mixed to prepare 32 pools (N1-N24 and N29-N36), and 13 D. silvarum DNA samples were mixed for 4 pools (N25-N28).All pooled DNA samples were sent to Biomarker Technologies Co., Ltd.(Beijing, China) for highthroughput sequencing of the 16S rDNA V3-V4 region on an Illumina NovaSeq 6,000 platform (Illumina, San Diego, CA, United States) with a sequencing depth of 80,000.The bioinformatics analysis of raw data was performed with the aid of the BMK Cloud (Biomarker Technologies Co., Ltd., Beijing, China).Raw data were filtered according to the quality of a single nucleotide using Trimmomatic (version 0.33) (16).Primer sequences were identified and removed using Cutadapt (version 1.9.1) (17).Clean reads obtained from the previous steps were assembled using USEARCH (version 10) (18) and followed by chimera removal using UCHIME (version 8.1) (19).Effective reads were generated and used for subsequent analyses.Sequences with similarity ≥97% were clustered into the same operational taxonomic unit (OTU) by USEARCH (version 10).Taxonomic annotation of the OTUs was performed based on the Naive Bayes classifier in QIIME2 (20) using the

Polymerase chain reaction
According to the results of high-throughput sequencing and tickborne bacteria prevalently detected in Inner Mongolia, the presence of Borrelia burgdorferi, B. miyamotoi, Anaplasma phagocytophilum, spotted fever group rickettsia, Ehrlichia chaffeensis, and Coxiella burnetii were tested in ticks and human blood samples using nested PCR or traditional PCR.The PCR primer sequences are listed in Table 1.PCR was performed using a PCR system (SensoQuest, Germany).The PCR system was totally 25 μL, including 12.5 μL Premix Taq™ (TaKaRa), 1 μL upper primer (10 μm), 1 μL lower primer (10 μm), 5 μL each individual DNA sample and 5.5 μL ddH 2 O. Positive (recombinant plasmid containing the target gene), negative (water), and extraction controls were used for PCR amplification in each PCR experiment.For nested PCR, 1 μL of the primary PCR production was used as the template for the second round.All amplified products were electrophoresed on a 1.5% agarose gel and the positive products were purified and sequenced.The acquired nucleotide sequences were submitted to National Center for Biotechnology Information Genbank and compared with available sequences deposited using the Basic Local Assignment Search Tool. 1  Additionally, human blood samples that were positive for SFGR based 1 https://blast.ncbi.nlm.nih.gov/Blast.cgi on nested PCR were tested using qPCR targeting gltA gene.qPCR was performed using a Probe qPCR mix (Premix Ex Taq™,TaKaRa) on a LightCycler 480 System (Roche Diagnostics, United States).

Phylogenetic analysis
Reference sequences of target genes downloaded from GenBank were aligned with the positive amplicon sequences using the ClustalW method with default parameters in MEGA11.0.One of the same nucleotide sequences detected in a tick or blood sample was selected as the representative sequence.Phylogenetic analysis was performed using the neighbor-joining algorithm, and bootstrap values were set for 1,000 replicates.

Serological tests
Based on the results of the high-throughput sequencing and PCR assays, SFGR and B. burgdorferi were the dominant tick-borne bacteria in Arxan.All serum samples were tested anti-SFGR by ELISA and anti-B.burgdorferi using IFA and WB.ELISA kits acquired from Enzyme Immune Industrial Co., Ltd.(Jiangsu, China) were used to detect anti-SFGR IgG and IgM according to the manufacturer's instructions.The cutoff value was the mean optical density (OD) of the negative control wells plus 0.15.Samples with OD values less than the cut-off value were considered negative.Samples with OD values greater than or equal to the cut-off value were considered positive.
IFA and WB were performed as previously described (29).First, all the serum samples were tested for anti-B.burgdorferi by IFA.A titer of ≥1/128 for IgG or 1:64 for IgM was considered positive.For IgG, samples that were positive at a titer of 1/64 but presented a weak Map of the tick sampling sites in Arxan, Inner Mongolia, China.

Statistical analysis
Statistical analysis was performed using Pearson's χ 2 test and p < 0.05 was considered significant.

Identification of ticks
A total of 295 adult hard ticks were identified as Ixodes persulcatus (n = 282) and Dermacentor silvarum (n = 13) based on morphological characteristics and mitochondrial 16S rDNA sequences.The nucleotide sequence data of mitochondrial 16S rDNA in this study will appear in GenBank under the following accession numbers: OR841358-OR841362 for I. persulcatus, OR841363-OR841367 for D. silvarum.

The overview of high-throughput sequencing
A total of 2,413,890 effective reads (range 69,960-76,736 reads, average 75434.06reads) of high quality (Q20 > 98.94%, shown in Additional file 1: Supplementary Table S1) were obtained from 32 pooled I. persulcatus samples.Subsequently, the effective reads were clustered into 1,445 Operational Taxonomic Units (OTU), which belonged to 32 phyla, 79 classes, 191 orders, 348 families, and 594 genera.Moreover, 303,375 effective reads were acquired from four pooled DNA samples of D. silvarum (range 75,414-76,021 reads, average 75843.75reads), all of which were of high quality (Q20 > 98.93%, shown in Additional file 1: Supplementary Table S1), and 1,430 OTUs were generated, belonging to 31 phyla, 78 classes, 190 orders, 346 families, and 588 genera.Alpha diversity reflects species richness and diversity of individual samples.The Ace and Chao1 indices, which were calculated to measure species richness, were more than 1068.34 and 1098.66,respectively, in each pool (Additional file 2: Supplementary Table S2).Shannon and Simpson indices were used to measure species diversity, which were affected by species abundance and community evenness in the sample community.With the same species abundance, the greater the uniformity of each species in the community, the larger the Shannon and Simpson indices values, indicating higher species diversity in the sample.Shannon diversity rarefaction (Figure 2) reflects the microbial diversity of each sample at different sequencing amounts, and was used to determine whether the sequencing depth was sufficient.All pools were highly diverse, with Simpson indices more than 0.85 except for N23.The species accumulation curve reflects the relationship between the sample size and the number of annotated species, and can be used to determine whether the sample size was sufficient.The species accumulation curve of I. persulcatus at the genus level is shown in Figure 3.The number of D. silvarum pooled DNA samples was insufficient for the analysis of the species cumulative curve.

Species annotation and taxonomic analysis
Taxonomic assignment revealed that the relative abundances of the genera were highly variable in the microbiome of I. persulcatus (Figure 4).At least 594 genera were presented in I. persulcatus, including Rickettsia, Acinetobacter, Methylotenera, Methylomonas, Psychrobacter.Rickettsia was the most common genus, with a relative abundance >22.61% in the seven pools (Additional file 3: Supplementary Table S3).Borrelia, with a relative abundance of 0.15-0.35%,was detected in three pools and further identified as B. miyamotoi.Likewise, Coxiella (0.01-0.02%) was identified in five pools and further confirmed to be Coxiella endosymbiont.Anaplasma (0.01%) was present in two pools, but species-level annotation was lacking (Additional file 4, Supplementary Table S4).
Similarly, Rickettsia was the most abundant genus in D. silvarum (Additional file 3: Supplementary Table S3).Coxiella was identified in each D. silvarum pool and confirmed to be Coxiella endosymbiont at the species level (Additional file 4: Supplementary Table S4), whereas Borrelia was not detected in D. silvarum.

FIGURE 2
Shannon index rarefaction curve of 36 pools of I. persulcatus and D. silvarum.When the curve tends to be flat, it means that the sequencing data are sufficient for analysis, and the Shannon index will not increase with increasing sequencing data.
In addition, the sequence (H-C74) detected in a blood sample fell into the same branch as strain Namibia belonging to genomic group IV (CP007555.1),C. burnetii from China (KR697576.1)(Figure 9).

Co-infection of pathogenic bacteria in ticks
In the present study, 27 I. persulcatus samples (27.00%, 27/100) were found to be co-infected with more than one bacterium (Table 3).Dual co-infection with B. garinii and Candidatus R. tarasevichiae (13.00%, 13/100) was most frequent in I. persulcatus.There was no co-infection in the participants according to PCR results.Species accumulation curve at genus level of 32 pools of I. persulcatus.Red boxes represent the accumulation curve of detected species numbers, and green boxes form the common curve of detected species numbers.When the curve tends to be flat, it means that the sample size is sufficient for the analysis.

Seroprevalence of SFGR and Borrelia burgdorferi
Among the 245 serum samples, 14 were positive for anti-SFGR by ELISA, of which four were only positive for IgG and 10 were positive for IgM and IgG (Table 4).Both IgG and IgM antibodies against SFGR were negative in those 2 individuals who were detected positive for R. slovaca using PCR.Forty serum samples were seropositive for B. burgdorferi using IFA, and positive samples were further detected by WB; 33 of them were positive for anti-B.burgdorferi IgG (Table 4).The representative image of positive western blotting reactions was displayed in  Additional file 5 (Supplementary Figure S1).Two serum samples were positive for both anti-B.burgdorferi IgG and anti-SFGR IgG.No significant difference in the seroprevalence of antibodies against SFGR and B. burgdorferi according to sex, age, or workplace was observed (Table 5).

Discussion
Ticks are recognized as one of the most important vectors of a wide variety of diseases in humans and animals.In recent years, an increasing number of tick-borne infections have stimulated the investigation of ticks and tick-borne pathogens (31,32).In this study, we applied 16S rDNA V3-V4 region high-throughput sequencing combined with species-specific PCR to detect tick-borne bacteria in free-living ticks and humans at high risk of tick-borne diseases in Arxan, Inner Mongolia, China.Subsequently, the seroprevalence of antibodies against SFGR and B. burgdorferi was determined to understand the infection rate of major tick-borne bacteria in the participants.
In this survey, the collected free-living ticks were mainly I. persulcatus with a small number of D. silvarum.Arxan City is located in northeastern China, which is mainly covered by coniferous and broad-leaved forests and is characterized by strong seasonality in temperature, providing a favorable habitat for I. persulcatus.The natural habitats of D. silvarum were characterized by middle to high elevations, shrub grasslands, and low precipitation during the driest month (2).In addition, we collected ticks in May 2020 and May 2021.The peak for I. persulcatus adults occurred in late May (33), while D. silvarum peaked in mid-April (34).Due to the limitation of the sampling season, sampling sites, and the number of ticks collected, only a small number of D. silvarum were collected.
High-throughput sequencing of 16S rDNA has been widely used to study microbial communities in ticks.For example, Runlsi et al. identified 189 genera in Haemaphysalis longicornis through high-throughput sequencing at the Illumina HiSeq platform, including Anaplasma, Rickettsia, and Ehrlichia (35).Jun et al. used high-throughput sequencing to study the microbial diversity of D. nuttalli and revealed the presence and relative abundance of the bacterial genera Rickettsia, Anaplasma and Coxiella (8).In the present study, at least 32 phyla, 79 classes, 191 orders, 348 families, Phylogenetic analysis for ompA gene of SFGR.The sequences clustered as Candidatus R. tarasevichiae were marked as "•," R. raoultii were marked as "▲," R. slovaca were marked as "◆."  (41).In the present study, R. slovaca was detected in two participants using nested PCR targeting ompA gene.This is the first study to report the presence of R. slovaca has been found in humans in Inner Mongolia, China.
Rickettsia raoultii was first detected in Dermacentor ticks collected in Russia in 1999 (42) and then isolated from Dermacentor ticks in France, and named in 2008 (43).Similar to R. slovaca, R. raoultii was more common among Dermacentor ticks.Other hard ticks infected with R. raoultii have been reported by several groups in recent years, including the Haemaphysalis, Hyalomma, and Ixodes ticks (44)(45)(46).In this study, R. raoultii was detected in nine D. silvarum using nested PCR.
Candidatus R. tarasevichiae, R. slovaca, and R. raoultii were all pathogenic genotypes of spotted fever group Rickettsia.The SFG patients present with rashes, eschar, fever, fatigue, anorexia, nausea, and local lymph node enlargement.A few patients had neurological manifestations such as coma, neck stiffness, and Kernig's sign (47).Combined with the presence of anti-SFGR antibodies among participants, public health workers in the area need to be aware of the risk of SFGR infection; regularly conduct education campaigns on tick bite prevention; and monitor populations, ticks, and animal hosts.Phylogenetic analysis for 5S-23S rRNA intergenic spacer region of B. burgdorferi sensu lato.The sequences clustered as B. garinii were marked as "•," B. afzelii were marked as "▲." populations in Inner Mongolia (15).Human Lyme disease generally occurs in stages, from the early localized stage of erythema migrans, fatigue, chills, and fever, to a late disseminated stage of intermittent bouts of arthritis with severe joint pain and swelling and neurological symptoms (29).Thus, Lyme disease should be considered by clinicians with the forestry workers who had correlated symptoms with Lyme disease in Arxan.
B. miyamotoi, initially identified and isolated in Japan in 1994 (50), is the causative agent of B. miyamotoi disease with generalized flu-like symptoms.Human cases of B. miyamotoi infection have been reported in European countries (51,52), the United States (53,54), Australia (55), and Japan (56).A survey of B. miyamotoi in Inner Mongolia suggested that 2.6% of I. persulcatus carried B. miyamotoi and 1.7% of patients bitten by ticks were infected with B. miyamotoi (57).In this study, B. miyamotoi which belongs to the Siberian type, was detected in 7% of I. persulcatus.This is the first report of this bacterium that has been detected in ticks from Arxan, indicating that local forestry populations are at risk of B. miyamotoi infection.
Coxiella burnetii, the causative agent of Q fever, is transmitted to humans via inhalation of infected aerosols or tick bites (58).In our study, C. burnetii was detected in one participant, indicating that extensive and in-depth monitoring of C. burnetii in ticks and animal hosts should be conducted to identify the risk of infection in the local population.
Ticks can acquire multiple pathogenic species during blood feeding on their vertebrate hosts, and humans may be infected by more than one pathogen carried by ticks.In this study, Candidatus R. tarasevichiae and B. garinii were the dominant pathogens detected in I. persulcatus, which, to some extent, explains why the rate of co-infection with B. burgdorferi and SFGR was the highest in I. persulcatus.Two participants tested positive for antibodies against both B. burgdorferi and SFGR, indicating that cases of coinfection with B. burgdorferi and SFGR were present in Arxan, Inner Mongolia.In recent years, concurrent infections with multiple tick-borne agents have been reported, including co-infection with B. burgdorferi s.l and A. phagocytophilum (59), Babesia microti and B. burgdorferi s.l (60), SFGR and severe fever with thrombocytopenia syndrome virus (61), etc. Pathogens may behave synergistically, indifferently, or antagonistically within common human hosts, thus prolonging the duration of symptoms and modulating disease severity (31).Additionally, the presence of concurrent infections may be neglected  Phylogenetic analysis for IS1111 gene of C. burnetii.The sequence clustered as C. burnetii was marked as "•."

Publisher's note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers.Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

FIGURE 4
FIGURE 4Relative abundances of top 10 microorganisms at the genus level in pooled I. persulcatus DNA samples.

FIGURE 5
FIGURE 5 were in pooled DNA samples of I. persulcatus, and the Ace and Chao1 indices were more than 1068.34 and 1098.66,respectively, in each pool indicating that all pools were of highly species richness.Simpson indices were more than 0.85 except for N23 indicating that almost all pools were highly diverse.Of 594 genera, Rickettsia, Anaplasma, Coxiella, and Borrelia, were detected in pooled DNA samples of I. persulcatus through high-throughput sequencing of the 16S rDNA V3-V4 region, Rickettsia was recognized as the most common genus.Candidatus R. tarasevichiae and three Borrelia species (B.garinii, B. afzelii and B. miyamotoi) were identified in I. persulcatus by nested PCR, whereas only R. raoultii was detected in D. silvarum.Candidatus R. tarasevichiae was first detected in I. persulcatus collected from the southern Urals and Siberia in 2003 (36) and was then found in Haemaphysalis japonica and D. silvarum from the Russian Far East (37, 38).Human cases of Candidatus R. tarasevichiae infection were first identified using laboratory molecular testing in eastern China in 2012 (39).In the present study, Candidatus R. tarasevichiae was detected in 89.00% of I. persulcatus, which was considered the dominant genotype of SFGR carried by I. persulcatus in Arxan.Rickettsia solvaca was first isolated from D. marginatus ticks in Slovakia in 1968, and subsequent investigations reported that R. solvaca is widely distributed in Dermacentor ticks (40).Tian et al. firstly detected R. slovaca in D. silvarum collected from Xinjiang Autonomous Region, China in 2012

FIGURE 8
FIGURE 8Phylogenetic analysis for glpQ gene of B. miyamotoi.The sequence clustered as B. miyamotoi was marked as "•."

FIGURE 9
FIGURE 9 The median age of all participants was 53.00 years (range 23-60 years).All the participants provided written informed consent to participate in this study.Sera were isolated from non-anticoagulant blood samples and stored at −20°C for serological tests.Anticoagulated blood samples were used for DNA extraction.

TABLE 1
Target genes and primer sequences used for PCR.

TABLE 2
Prevalence of tick-borne bacteria in ticks and humans.

TABLE 3
Prevalence of co-infection with tick-borne bacteria in I. persulcatus.BacteriaNumber of ticks positive for co-infection (%)

TABLE 4
Seroprevalence of antibodies against B. burgdorferi and SFGR.

TABLE 5
Differences in seroprevalence of antibodies against B. burgdorferi and SFGR according to sex, age, and workplace.