Characterization of the bacterial microbiome of Rhipicephalus (Boophilus) microplus collected from Pecari tajacu “Sajino” Madre de Dios, Peru

Ticks are arthropods that can host and transmit pathogens to wild animals, domestic animals, and even humans. The bacterial microbiome of adult (males and females) and nymph Rhipicephalus microplus ticks collected from a collared peccary, Pecari tajacu, captured in the rural area of Botijón Village in the Amazon region of Madre de Dios, Peru, was evaluated using metagenomics. The Chao1 and Shannon–Weaver analyses indicated greater bacterial richness and diversity in female ticks (GARH; 375–4.15) and nymph ticks (GARN; 332–4.75) compared to that in male ticks (GARM; 215–3.20). Taxonomic analyses identified 185 operational taxonomic units representing 147 bacterial genera. Of the 25 most prevalent genera, Salmonella (17.5%) and Vibrio (15.0%) showed the highest relative abundance followed by several other potentially pathogenic genera, such as Paracoccus (7.8%), Staphylococcus (6.8%), Pseudomonas (6.6%), Corynebacterium (5.0%), Cloacibacterium (3.6%), and Acinetobacter (2.5%). In total, 19.7% of the detected genera are shared by GARH, GARM, and GARN, and they can be considered as the core microbiome of R. microplus. To the best of our knowledge, this study is the first to characterize the microbiome of ticks collected from P. tajacu and to report the presence of Salmonella and Vibrio in R. microplus. The pathogenic potential and the role of these bacteria in the physiology of R. microplus should be further investigated due to the possible implications for public health and animal health in populations neighboring the habitat of P. tajacu.


Results
Ticks collected from P. tajacu. Taxonomic identification indicated that all the collected ticks in Madre de Dios ( Fig. 1) belong to R. microplus 12 . Statistical values and diversity in the R. microplus microbiome. Microbiome analysis using the 16s-515F/16s-806R primers and amplicon sequencing on Ion Torrent PGM (Ion Personal Genome Machine System, THERMO FISHER SCIENTIFIC) generated a total of 117,192 raw reads (39,604 average) from the three analyzed samples [13][14][15] (Table 1). After rigorous data curation, 55,805 high-quality sequences were retained with an average of 20,462 sequences per sample and an average length of 150 bp 16,17 . The maximum number of filtered sequences (26,549) was obtained from the female tick sample, which exceeded those found in male www.nature.com/scientificreports/ and nymph samples by 164.7% and 204.8%, respectively 18 . These sequences were assigned to 1075 total unique sequences corresponding to 185 abundant (< 0.005%) OTUs based on a > 97% identity cutoff for bacterial 16S rRNA genes 18 . At the individual sample level, the microbiome from nymphs surpassed that from females and males (221,195, and 148 OTUs, respectively). Those OTUs were mainly identified as prokaryotes (99.89%) and to a lesser extent as unknown sequences (0.11%). At the taxonomic level, a total of 147 genera distributed in 99 families, 59 orders, 30 classes, and 12 phyla were detected. The SW index reflects the specific diversity of each sample, whose value increases as the number of different OTUs increases. In this study, the microbiome obtained from nymph tick samples showed a higher SW index than the female and male microbiomes. On the other hand, Chao1, the index that evaluates specific richness, showed that the number of expected OTUs decreased from 375 in GARH to 332 in GARN and 215 in GARM after the standardization of the sample size to 14,000 sequences. Statistical analyses of variance of the SW and Chao1 indexes in the GARH, GARM, and GARN samples showed significant differences (P < 0.01) 19-21 . Composition of the core and shared and individual microbiome from R. microplus. The comparative analysis of the composition of the microbiota from GARH, GARM, and GARN revealed that 19.7% out of the 147 genera found in R. microplus were common to the three groups. This shared community is considered as the core microbiota ( Table 2). The percentages showed a decreasing proportionality in GARH, GARN, and GARM in relation to the non-shared bacterial genera. A higher percentage of shared microbiota was observed between GARH and GARN (9.5%) compared to that between GARM and GARH (3.4%) and between GARM and GARN (1.4%).
Microbiota according to R. microplus sex and stage. Regarding the abundance of bacterial genera in R. microplus, Salmonella was the most abundant genus in GARM, while Vibrio was the most abundant genus in Table 1. Statistical summary of the microbiota from Rhipicephalus microplus. *Significant differences (P < 0.01). www.nature.com/scientificreports/ GARH and GARN, and Paracoccus was the second most abundant genus in GARH. On average, these were the most abundant genera in R. microplus, 17.5%, 15%, and 7.8%, respectively (Fig. 2).

Discussion
The richness and diversity indexes revealed that the microbiota present in GARH and GARN exhibit greater bacterial genera diversity and richness than the microbiota in GARM. This is in agreement with previous studies on R. microplus that were collected from cattle 9 . Previous studies in male and female ticks of Ixodes ovatus, I. persulcatus, and Amblyomma variegatum have shown differentiated microbiome profiles both at the taxonomic and functional levels between sexes of the same tick species 22 .
A metagenomic study showed that the microbiome profile in ticks is related to metabolic processes and that their resilience and adaptability to the environment is related to their sex 22 . In addition, geographical location, temperature, humidity, species, sex, anatomical location, and type of diet have been shown to affect the microbiome of ticks [23][24][25][26][27][28] . In our study, although ticks were of the same species and were collected from the same host, significant differences were found in bacterial diversity and richness related to the sex and developmental stage of ticks.
Among the 147 different genera identified, the core microbiome that included the majority of the most prevalent genera stood out. Several of the identified genera within the core microbiome are known to be human pathogens (i.e., Salmonella, Vibrio, Paracoccus, Staphylococcus, Pseudomonas, Corynebacterium, Cloacibacterium, and Acinetobacter). In addition, a greater bacterial microbiome was shared between nymph and female ticks [14 (9.5%)] compared to that shared between male and female ticks [5 (3.4%)]. We suggest that these differences have a behavioral origin. Thus, female and nymph ticks are more prone to remain on the same host, whose microbiota impact on the tick gut microbiome, while male ticks frequently change hosts 22 . This hypothesis is supported by studies on other genera that reported higher relative abundance and alpha diversity in female ticks than in male ticks 22 . Additionally, it is necessary to consider that the role of nuclei bacterial genera and the species included in these may present different roles as pathogens or symbionts depending on whether they are found in the arthropod or in the vertebrate that hosts the arthropod.
The most prevalent genus among the three groups of ticks was identified as Salmonella, whose members cause gastrointestinal tract infection and dysentery and can lead to serious clinical conditions, especially in children 29 . The genus Vibrio, the second in abundance (15.6%), represents a finding of great interest as, to the best of our knowledge, this is the first study showing its presence in R. microplus. The genus Vibrio is a common commensal of aquatic arthropods and has a remarkable capacity for adaptation to the environment 30,31 . Its presence evinces the adaptation of this genus to the gastrointestinal system of R. microplus, which inhabits a jungle ecosystem. www.nature.com/scientificreports/ Many Vibrio are opportunistic pathogens of both arthropods and humans. Therefore, studying the virulence of the identified species is essential [30][31][32] . Paracoccus, the third most abundant genus (6.97%), is a coccobacillary bacterium that is typically present in a wide range of ecosystems 33 . Staphylococcus, with a prevalence of 6.63%, is mainly related to infections in soft tissues and has been previously reported in the gut of R. microplus and with a high prevalence in female Amblyomma variegatum 9,22 . Pseudomonas showed an abundance of 5.87% in R. microplus. In previous studies, the presence of this bacterial genus in R. microplus and in male Amblyomma variegatum with a high prevalence has been reported 9,22 . Pseudomonas has been suggested to be involved in the infection of soft tissues, including the tissues of the respiratory system 34,35 . The presence of Corynebacterium, with an abundance of 5.87%, is important because some Corynebacterium species produce the diphtheria toxin or can cause osteomyelitis 36 . In addition, this genus has been previously identified in eggs and male adults of R. microplus 9 . Cloacibacterium, with a prevalence of 2.93% in R. microplus, are gram-negative bacteria that proliferate in aqueous environments with high content of organic matter 37 . Acinetobacter, with an abundance of 2.53%, has been reported in a metagenomic study in I. persulcatus, I. pavlovskyi, and Dermacentor reticulatus 38 . Sphingomonas, the ninth most abundant genus (2.47%), includes non-fermenting and strictly aerobic gram-negative bacteria. Some species, such as S. paucimobilis and S. wittichii, can cause infections in immunocompromised patients 39,40 .
In contrast to the bacterial microbiome relevant to human health identified in our study, a previous study on bacterial diversity in R. microplus collected from cattle identified Ehrlichia sp., Coxiella sp., and Bartonella sp. 41 . This indicates that the bacterial microbiome would also depend on the host parasitized by the ticks. Some bacteria, such as Leptospira interrogans, Mycobacterium, Salmonella, Clostridium, and Pasteurella, and tick genera, such as Haemaphysalis, Dermacentor, and Amblyoma, have been identified in the genus Pecari 42-44 . In our case, R. microplus, a tick that mainly parasitizes cattle 45 , was found in P. tajacu (sajino). P. tajacu was possibly tick infected due to the proximity of Botijón Village, where livestock farming is practiced. This highlights that ticks can infect cattle, P. tajacu, and humans, with the potential risks of pathogen transmission that this implies.
Regarding the role of bacteria in ticks, note that nonpathogenic microorganisms present in ticks could cause infections in humans and other animals. For example, ecological studies have shown that Rickettsia, Francisella, and Coxiella, which are considered vertebrate pathogens, can change their pathogenic role and have a mutualistic and symbiotic relationship with ticks 1 . Therefore, studying the interaction between the bacterial microbiota and ticks is of utmost importance for the control of pathogens and the development of the arthropod 1 . Coxiella sp. infects at least two-third of the ticks and is important for the survival of Amblyomma americanum and Rhipicephalus sp. 46,47 . Nonetheless, it has not been found in our study. Coxiella sp. and Francisella sp. are linked to the synthesis of vitamins necessary for the survival of ticks [48][49][50] . Likewise, other symbiotic bacteria, such as Francisella, Rickettsia, and Rickettsiella, have been reported 46 , with Rickettsia sp. and Coxiella sp. having become strict endosymbionts 1 . According to previous studies, the endosymbiont bacteria of a species of tick vary depending on the ecology and the number of ticks studied, for example, although in the case of Coxiella it was previously described R. microplus collected from cattle, previous studies that have demonstrated that the infection rates by Coxiella in R. microplus ticks are highly variable. A 2016 study evaluated R. microplus from Brazil, and found that only approximately 37% of the samples contained Coxiella 51 . In 2015 a study evaluated Coxiella in many species of ticks without finding the bacteria, one of the species of ticks evaluated was 3 R. microplus samples from Benin (west Africa) and did not find any Coxiella 52 . Therefore, the importance of our study is the finding of the new microbiome of R. microplus collected from Pecari tajacu.
The small number of ticks was justified by the fact that R. microplus ticks are not very common on the wild host Pecari tajacu; therefore, we could not collect a larger sample of ticks. On the other hand, we found interesting to test these ticks because We wanted to search for the microbiota of an exotic tick (R. microplus) infesting a mammal species native to the Amazonia (Pecari tajacu). Again, even though our sample was small, we have to highlight the interesting results We have found from these ticks.
Among the limitations of our study is the bacterial microbiome found in 5 females, 5 males and 2 nymphs of ticks collected from P. tajacu, which implies a bacterial microbiome representative of a specific circumstance and ecology. Therefore, studies with a greater number of samples could show a greater diversity of species and different percentages of bacterial abundance.

Conclusion
In this study, we found a high bacterial diversity in female, male, and nymph R. microplus collected from P. tajacu. The greatest bacterial diversity and richness was found in females and nymph ticks compared to male ticks. The most frequent bacterial genera were Salmonella, Vibrio, and Paracoccus. This is the first bacterial metagenomic study performed in R. microplus collected from P. tajacu in the Peruvian jungle, and the presence of Vibrio is highlighted. This study lays the foundations for future studies on the importance of the role of the identified bacteria on arthropods and animal and human health.

Material and methods
Ethical aspects. This study was approved by the Oficina de Salud Pública y Medio Ambiente del Consejo Regional de Madre de Dios (Office of Public Health and Environment of the Regional Council Madre de Dios), Peru. Laboratory procedures for bacterial identification were conducted in accordance with the international guidelines for the use of animals in research and the standards of the Comité de Cuidado y Uso de Animales del Área de Investigación en Salud de la Junta del Consejo Regional de Madre de Dios (Animal Care and Use Committee of the Health Research Area of the Madre de Dios Regional Council Board). The study was carried out in compliance with the ARRIVE guidelines.  (Fig. 1). The collection site corresponds to a forest area where hunting of wild animals is allowed. The average annual rainfall in the study area is 1600 mm 3 , and the average annual temperature is 25 °C. The area is in the tropical wet forest zone. During sample collection, the weather was hot and humid.

Sample collection.
A wild male of P. tajacu (sajino) was captured in Botijón Village in June 2012. The ticks from its abdominal region were collected 3 h after its sacrifice using forceps and were individually placed in 2 ml cryovials containing 96% ethyl alcohol. Cryovials were labeled with an identification code for the sampling site and the animal from which the sample was collected. Five male ticks, five female ticks, and two nymph ticks were identified. On sterile plates ticks were washed for 15 min in a solution 0.9% isotonic sterile sodium chloride saline followed by 15 min in a solution of 96% ethanol to remove surface contaminants. Excess solution was absorbed and ticks were air-dried prior to manipulation under sterile conditions. Each tick was individually cut in half lengthwise using sterile scalpels number 15.
Taxonomic classification. Ticks were identified using taxonomic keys 12  Total DNA extractions were analyzed by spectrophotometry (NANODROP EPPENDORF), and the samples with sufficient quality and quantity were shipped to MR DNA (Shallowater, TX, USA) and sequenced on the PGMplatform (Ion Personal Genome Machine System, THERMO FISHER SCIENTIFIC). Metagenomic analysis was performed on the PCR amplification products of the V4 hypervariable region of the 16S rRNA gene using the 515F/806R primers 15 . Analysis and processing of metagenomic data. The sequences generated by Ion Torrent were analyzed with QIIME v1.9.1 15 , where the initial sequences were processed based on filtering of barcodes ≤ 6 bp, Q25 quality scores, 150 bp sequence length, and chimera detection using usearch61 16,17 . High-quality sequences were assigned to operational taxonomic units (OTUs) with a 97% identity cutoff for bacteria. The final OTUs were classified taxonomically using the High Quality Ribosomal RNA Databases "SILVA" v132 database (https:// www. arb-silva. de/). Likewise, unrepresentative OTUs ≤ 0.005% were filtered during analysis 18 .