Characterization of bacterial communities in ticks parasitizing cattle in a touristic location in southwestern China

The purpose of this study was to investigate tick species around Mount Fanjing and analyze bacterial communities in two species – Rhipicephalus microplus and Haemaphysalis longicornis – parasitizing cattle in Tongren, Guizhou province, Southwest China, using high-throughput sequencing methods. In April 2019, ticks were collected from five sites in Jiangkou County, Yinjiang County, and Songtao County. In total, 296 ticks were collected, comprising two genera and three species: H. longicornis, Haemaphysalis flava, and R. microplus. Rhipicephalus microplus was the most representative species (57.4%) within the collected group, being the dominant species in Tongren City, followed by H. longicornis (39.5%) and H. flava (3.0%). Beta-diversity analysis revealed differences in bacterial community composition among the tick species. The bacterial community structure of R. microplus collected in the three counties was highly similar. Chlorella and Bacillus were highly abundant in H. longicornis. Rickettsia was detected at high relative abundance in R. microplus but in low relative abundance in H. longicornis, suggesting that Rickettsia is more associated with R. microplus than with H. longicornis. More in-depth investigations are needed to determine the pathogenic risk of Rickettsia and its relationship with the host. This is the first survey on tick-borne bacterial communities in this area, which is of great significance for the prevention and control of tick-borne diseases locally. Supplementary Information The online version contains supplementary material available at 10.1007/s10493-023-00799-y.


Tick identification
Morphological features of ticks were identified based on taxonomic and morphological criteria (Teng and Jiang 1991;Chen and Yang 2021) using an SZX7 stereomicroscope (Olympus, Chaoyang, Beijing, China) and a Smartzoom 5 digital microscope (Zeiss, Pudong, Shanghai, China). Observed structures included the shape of basis capituli, porose area, palp, scutum, and peritreme, as well as the dentition formula of hypostome, the presence or absence of festoon, among other structures.

Sample grouping
Based on collection site, tick species, and tick size, representants of the same tick species originating from different collection sites within the same geographical area were pooled. All ticks were fully engorged. The total volume of each sample was approximately the size of a soy bean, and samples were divided into four groups (Table 1) with three replicates each. The negative control used ddH 2 O instead of sample DNA to exclude environmental and reagent contamination, and the rest of the experimental conditions and procedures were consistent with the samples.

Sample preparation
Each sample was washed 3× with 75% ethanol (Lircon, Dezhou, Shandong province, China), and then once with 1× PBS buffer (Life Technologies, Grand Island, NY, USA), and placed in a 2-mL centrifuge tube to prepare for nucleic acid extraction.

Library quality control and sequencing
Library quality was determined by electrophoresis on 2% agarose gels. DNA fragments of approximately 400 bp were recovered using magnetic beads. The quality and concentration of prepared libraries were determined using Qubit v.3.0 fluorometer. High-throughput sequencing was performed using the Illumina MiSeq platform by Sangon Biotech (Shanghai). The company also performed a bioinformatic analysis for us.

Bioinformatics analysis
Raw sequence image data obtained by high-throughput sequencing were analyzed by base calling and transformed into raw sequenced reads, and saved as FASTQ files. Adapters were removed using cutadapt v.1.18, and paired-end reads were merged using PEAR v.0.9.8 using the overlapping method. Finally, quality control and filtering were carried out in each sample to obtain clean data. Operational taxonomic unit (OTU) clustering was performed on clean sequences, and chimera sequences were removed to obtain optimal OTU sequences, which were selected as representative sequences with ≥ 97% similarity. Then the generated OTUs were compared with the ribosomal database project (RDP, http://rdp.cme.msu.edu/ index.jsp). Sequence classification assignments were performed using the RDP classifier, which was based on Bergey's taxonomy, and the Naive Bayesian assignment algorithm was used to calculate the probability value of each sequence assigned to various hierarchical levels.
Six metrics (indices) were used for determining alpha-diversity analysis: Shannon, Chao, ACE, Simpson, Shannoneven, and Coverage. The Shannon and Simpson indices indicate the diversity of bacterial communities in the sample, and the Chao and ACE indices indicate the abundance of bacterial communities. The Shannoneven index was used to reflect the evenness of bacterial communities, and the Coverage index reflects the coverage of each sample library.
Data analysis was performed in the statistical language R (v.3.2). Hierarchical clustering was based on the Bray-Curtis distance algorithm, and the package 'ape' (v.5.3) of R was used to construct dendrograms.

16S rDNA sequencing
The effective number of reads (Table 3) in samples was within the range of 30,000-65,000, and the average length was approximately 400 bp. In total, 904 OTUs were identified in all samples. A rarefaction curve was constructed with the number of reads in the horizontal axis and the number of OTUs in the vertical axis (Fig. 1). The observed number of OTUs increased with sequencing depth, and the curve tended to become flat towards its end, indi-  Table 3 Number and length of sequences of each sample in Tongren City, Guizhou province. See Table 1 for details on the sample codes cating that sequencing depth was satisfactory to cover all bacterial species in the sample. Thus, sequencing data could likely reflect the features of the tick-borne bacterial community in the investigated geographical area.

Alpha diversity
Collectively, the obtained data revealed that the diversity and abundance of microbial communities in each sample differed. Diversity was higher in samples JK-1 (R. microplus) and ST-4 (H. longicornis) and lower in samples JK-2, ST-1, and YJ-1 (all three R. microplus).
Abundance was highest in sample JK-1 and lowest in sample JK-3 (R. microplus). Community distribution in samples JK-1 and ST-4 was more uniform than in the other samples.
Coverage indices of all tested samples were > 99.7%, which indicated the reliability of our sequencing data pertaining the microbial community and low probability of sequences not being detected (Table 4).  Table 1 for details on the sample codes

Beta diversity
Beta-diversity analysis was conducted to determine the dissimilarity of bacterial communities in ticks evaluated in the study. Two main branches (Fig. 5) were identified in dendrograms: R. microplus samples were found in a branch regardless of the site of collection, and H. longicornis samples fell in other branches. Sample ST-5 differed from the other 11 samples in the corresponding branch due to the presence of Bacillus (66.3%), Staphylococcus (11.4%), and very little Chlorobium (0.04%); sample ST-4 was similar to sample ST-6 as the abundance of Chlorobium was 50% and Rickettsia was found in low abundance in both samples, 2.4 and 0.05%, respectively.  Table 1 for details on the codes in the periphery

Discussion
The geographical distribution of the collected ticks was different: in sites i and ii (in Jiangkou County), the tick species were highly similar (mainly R. microplus); in site iii (Yinjiang County), R. microplus was the only species identified, and in sites iv and v (Songtao County) H. longicornis was the dominant species. The environment of the five sampling sites is similar and all locations are near Mount Fanjing, and they all have a subtropical humid monsoon climate. The average annual temperature is 16-17 ℃, the average temperature of the hottest month (July) is about 27 ℃, and that of the coldest month (January) is about 5 ℃. But sampling site v has some uniqueness: it has 230 ha of artificial pasture, stocked with cattle, sheep, rabbits, and other animals; the other sampling sites only stocked cattle. Perhaps this explains the difference in the dominant tick species. Rhipicephalus microplus is the main ectoparasite affecting livestock worldwide (Silva et al. 2016), whereas H. longicornis is a tick species widely distributed in China (Malik et al. 2019).

Fig. 3
Relative abundance of microbial genera in ticks. Genera whose relative abundance was < 1% in all samples were depicted as 'other'. See Table 1 for details on the sample codes In this study, using 16S rDNA sequencing analysis to explore the diversity of bacterial communities in collected ticks, the species Rickettsia, Chlorobium, and Bacillus were detected in all 12 samples. Rickettsia was detected at the highest relative abundance in R. microplus. Lim (2020) applied high-throughput sequencing to characterize the diversity of bacterial communities in ticks and also detected Rickettsia in high abundance in Dermacentor atrosignatus and Dermacentor compactus. Bacteria within the genus Rickettsia, including spotted fever group rickettsiae (SFGR), are mainly transmitted to animals and humans  Table 1 for details on the sample codes by tick bites; the genus Rickettsia has at least 30 species distributed worldwide, of which 21 are considered pathogens (Satjanadumrong et al. 2019). SFGR are known to cause a variety of natural focal diseases, including the Rocky Mountain spotted fever (RMSF) (Zazueta et al. 2021) in the Americas and the Mediterranean spotted fever in parts of Asia, Europe, and Africa (Satjanadumrong et al. 2019). In Europe, SFGR is chiefly transmitted by Dermacentor (Buczek et al. 2020), whereas in Asia, Dermacentor and Haemophysalis are most frequently associated with rickettsial carriage. SFGR is highly prevalent in northern China. Ten validated SFGR species have been discovered in ticks and vertebrate hosts . Nine samples (75%) evaluated in this study were shown to contain a high relative abundance of Rickettsia -they may be pathogens or endosymbionts, so we need further investigation to assess the possible risk posed by R. microplus.
In addition, seven samples were shown to contain a low abundance of Coxiella, ticks collected in Jiangkou County (samples JK-1, 2, and 3), Yinjiang County (YJ-1, 2, and 3), and Songtao County (ST-4). These findings differed from those of Guizzo et al. (2017), in which Coxiella accounted for 99% of the bacterial community in R. microplus eggs and 98.3% in R. microplus larvae. These differences may be attributed to the stage of development of ticks and the surveyed geographical location, but of course this requires confirmation. Bacteria within the genus Coxiella establish a symbiotic relationship with ticks and are able to infect ticks at all stages of their lifecycle (Ni et al. 2020). Coxiella burnetii is an obligate intracellular bacterium and the agent of Q fever in humans; Q fever is a zoonotic disease widely prevalent worldwide (except for New Zealand), characterized by high infectivity and long-term environmental persistence (Long et al. 2019;Klemmer et al. 2018). Ticks were not considered vectors of C. burnetii previously (Abdelkadir et al. 2019), but the results of  Table 1 for details on the sample codes a recent study support the link between ticks and Q fever, with the latter 3× more likely to occur where the former is found (Hussain et al. 2022). Ticks play an important role in the wild and peridomestic cycles of C. burnetii worldwide, having been isolated from at least 40 tick species within the Ixodidae and 14 tick species of the Argasidae (Bolaños-Rivero et al. 2017). Although animals infected with C. burnetii are often asymptomatic, sheep, goats, and cattle may experience abortion, premature birth, stillbirth, and weak offspring (Di Domenico et al. 2014), thus the bacterium can disseminate in the environment through birth products as well as the urine and milk of infected animals (Rodolakis et al. 2007). Moreover, as it is extremely resistant to desiccation and radiation, C. burnetti can persist in soil and other dry surfaces for long periods (Körner et al. 2021). Humans are very sensitive to C. burnetii, especially those who work with livestock and are exposed to birth products, infectious dust particles, contaminated wool, and highly infectious aerosols. Infections by C. burnetii are often occupation-related, being highly prevalent among slaughterhouse workers, livestock handlers, veterinarians, and farmers (Frangoulidis et al. 2021;Klemmer et al. 2018;Esmaeili et al. 2014).
In the present study, Proteobacteria were detected at the highest abundance in R. microplus, whereas Chlorobi and Firmicutes were at the highest abundance in H. longicornis, thus indicating that bacterial communities differ among tick species. Chlorobium was detected in ticks evaluated in the current study, which are bacteria that mainly live in water and are able to perform photosynthesis; the presence of this bacterium in H. longicornis may be a result of tick's water ingestion habits. For instance, Amblyomma americanum and Ixodes scapularis have been shown to be able to actively ingest liquid water from the environment to compensate for water loss occurring due to excretion processes (Maldonado-Ruiz et al. 2020Kim et al. 2017Kim et al. , 2019. Therefore, Chlorobium was likely ingested along with water by H. longicornis. Of course, the possibility that these bacteria originated from the surface of the exoskeleton and the host skin could not be ruled out, as it is quite difficult to completely remove surface contaminants from the tick exoskeleton. Interestingly, the results of the present study differed from previous studies conducted in other provinces which employed high-throughput sequencing technology. Zhang et al. (2019c) evaluated the bacterial communities of H. longicornis, in which Coxiella was shown to be the dominant genus, and Rickettsia was not detected. Xiang et al. (2017) studied the bacterial communities of saliva obtained from engorged adult R. microplus females in Hunan province and found that Proteobacteria was the dominant phylum, and Acinetobacter, Rickettsia, Escherichia, and Coxiella were the major genera. Zhang et al. (2021b) also detected a high abundance of Rickettsia in H. longicornis in Shandong province. At present, it is not possible to provide a comprehensive explanation of the changes in bacterial richness observed in the current study with those described in previous studies. Multiple factors can determine changes in tick microbiome, including tick species, sample collection site, presence of host blood, and degree of engorgement (Clow et al. 2018;Estrada-Peña et al. 2018;Swei and Kwan 2017;Van Treuren et al. 2015;Moreno et al. 2006). In addition, several bacterial species detected in the study, including Acinetobacter and Staphylococcus, are commonly detected in ticks, indicating that these bacteria may play a biological role in these hosts. Moreover, it is worth noting that species from the genera Rickettsia and Coxiella are common tick endosymbionts (Dall'Agnol et al. 2021;Lim et al. 2020;Maldonado-Ruiz et al. 2020). Most importantly, the high carrying rate of Rickettsia in ticks evaluated in the study sites towards the need to strengthen future investigations to identify vector competence and potential epidemiological risk of tick-borne disease in the surveyed area.
Finally, high-throughput 16S rDNA sequencing has become the standard method for microbial classification and identification, which enables the identification of bacteria found in very low abundance and/or non-cultivable states. However, this method has certain limitations, such as accuracy and comprehensiveness being associated with the reference database (Couper and Swei 2018). Considering only the V3-V4 hypervariable regions of the 16S rDNA gene, sequence length was likely insufficient to enable the identification of bacteria at the species level. 16S rDNA full-length sequencing technology yields sequences of approximately 1542 bp in length, including nine hypervariable regions and 10 conserved regions, which would allow for further classification and identification of bacterial communities at the species level (Dong et al. 2021). Unfortunately, the limited number of samples did not allow further investigation of the identity of these bacteria to establish whether these bacteria are pathogens or not. However, the high-throughput sequencing method employed in the present study enabled the successful exploration of the bacterial communities carried by ticks parasitizing cattle in Tongren, Guizhou province, and highlighted variation in bacterial communities found in R. microplus vs. H. longicornis, the most prevalent tick species in the surveyed areas. Collectively, the results discussed herein provide a scientific basis for strengthening prevention and control measures of ticks and tick-borne diseases in Guizhou province.

Conclusions
The present study constitutes a survey of tick distribution affecting cattle in Tongren, Guizhou province, China, as well as of the diversity and composition of microbial communities found in the two tick species most prevalent locally. Regional differences were found in the distribution of tick species as well as in bacterial species carried by these parasites. Future studies should comprise an in-depth analysis of tick microbiota composition at the species level and explore the role of the identified microorganisms in ticks.

Acknowledgements
The authors gratefully acknowledge Sangon Biotech (Shanghai) Co., Ltd for the technical support, Tongren Center for Disease Control and Prevention for its help in the sampling process, and farmers for their understanding and cooperation. The authors would like to thank TopEdit (www.topeditsci. com) for its linguistic assistance during the preparation of this manuscript.
Author contributions XYL, ZJZ and LWQ conceived and designed the experiment. XYL, and ZY performed laboratory work. YFX, LSJ and HY interpreted results. XYL and ZJZ compiled tables and figures. XYL and ZJZ wrote the first draft of the manuscript, and LWQ and LQY contributed to finalizing the paper. All authors have read and agreed to the published version of the manuscript.