Evidence for transovarial transmission of tick-borne rickettsiae circulating in Northern Mongolia

Transstadial transmission of tick-borne rickettsiae has been well documented. Few studies, however, have evaluated the role of transovarial transmission of tick-borne rickettsiae, particularly in nature within the host-vector ecosystem. This cross-sectional study aimed to understand the role of transovarial transmission of tick-borne rickettsiae among feeding ticks at different life stages. Tick eggs laid by engorged wild-caught adult female ticks were pooled and tested for Rickettsia spp. and Anaplasma/Ehrlichia spp. using molecular techniques, while adult fed ticks were tested individually. Additionally, larval and nymphal ticks were collected in the wild from small mammals, pooled and tested for Rickettsia spp. and Anaplasma/Ehrlichia spp. There were 38 fed adult and 618 larvae/nymphs (60 pools total) Dermacentor spp. ticks collected from livestock and rodents. All individual adult ticks and tick pools were positive for Rickettsia spp. While none of the larvae/nymphs were positive for Anaplasma/Ehrlichia spp., two adult fed ticks were positive. Rickettsia spp. DNA was detected in 91% (30/33) of the pooled eggs tested, and one pool of eggs tested positive for Anaplasma/Ehrlichia spp. Sequencing data revealed Rickettsia spp. shared ≥99% identity with R. raoultii ompA. Anaplasma/Ehrlichia spp. shared ≥89% identity with A. ovis 16S ribosomal RNA. This study identified potential transovarial transmission of Rickettsia spp. and Anaplasma spp. among D. nuttalli ticks. Additional studies are needed to further assess the proportion of transovarial transmission occurring in nature to better understand the burden and disease ecology of tick-borne rickettsiae in Mongolia.


Introduction
While significant effort has been directed to study tick-borne rickettsiae, they continue to be a global public health threat.Mongolia is a country known for its rich nomadic and pastoral culture, with populations of people who work very closely with their livestock in environments that are often densely populated with ticks.Additionally, ecotourism is a rapidly growing industry in Mongolia, placing international visitors at risk of exposure to tick-borne rickettsiae [1,2].This public health challenge is further complicated by a limited knowledge and understanding of tick and tick-borne rickettsiae ecology within Mongolia [2][3][4].
The mobility of ticks is restricted to questing and travelling via feeding on animals and humans [5].Tick-borne rickettsiae typically undergo transstadial transmission before being vectored by a competent tick host.However, depending on the tick species and the type of tick-borne rickettsiae, transovarial transmission may also occur [6].Research related to transovarial transmission has been particularly limited within the Asian and Eurasian regions of the world.
Tick-borne rickettsiae of most concern in the Asian and Eurasian regions of the world are Rickettsia spp.[21,24], Anaplasma spp.[12,14,16,25,27,32], and Ehrlichia spp.[17,25].These rickettsiae have been associated with small mammal reservoirs [6,17,19].Collectively, the objectives of this study were to further investigate the life cycle of tick-borne rickettsiae in locally occurring ticks; to examine the propensity of certain tick-borne rickettsiae to undergo potential transovarial transmission; and to evaluate the infection prevalence of tickborne rickettsiae infections from early life stage ticks throughout the Northern Mongolia region.

Methods
This cross-sectional study was designed to evaluate ticks at different life stages (Fig 1).First, engorged adult ticks were collected from livestock located in three soums (counties) within three aimags (provinces) (

Adult tick and egg collection
Handling procedures for livestock were conducted by trained veterinary staff prior to this study during animal care and were in accordance with the Mongolian Institute of Veterinary Medicine, Ulaanbaatar, Mongolia.Verbal consent was obtained from livestock owners at time of tick collection.Female adult fed ticks were collected from livestock at time of veterinary care of livestock and kept alive at room temperature in storage containers at the Laboratory of Arachno-Entomology and Protozoolgy, Institute of Veterinary Medicine in Ulaanbaatar, Mongolia.Moist cotton was placed near the ventilation of the containers to replicate environmental humidity conditions.Once female ticks laid eggs (between 2-7 days of incubation), both adult ticks and eggs were stored separately at -80˚C until DNA extraction was performed.The whole egg clutch was pooled and tested from each adult female tick.Mass of egg clutches ranged from 10 to 410 milligrams.Eggs and adult ticks were briefly rinsed with 70% ethanol in sterile 1 mL vials to remove contamination and then air dried on a sterile dish in preparation for processing [33,34].

Larval/nymphal tick collection
Trapping and handling procedures for small mammals were approved by the Duke University Institutional Animal Care and Use Committee (#A086-16-04) in accordance with the Mongolian Institute of Veterinary Medicine, Ulaanbaatar, Mongolia.At each location, live Tomahawk and Sherman traps were placed near holes where there were signs of recent small mammal habitation.All captured small mammals were sedated with ketamine (50 mg/Kg) and inspected for ticks.Ticks were stored in 70% ethanol at room temperature.Specimens were taxonomically identified to genus for larvae and nymphs and species for adults by a trained entomologist.Ticks were air dried and pooled based on life stage (larvae range 1-15; nymphs

Polymerase chain reaction and sequencing
All ticks and eggs were ground using a sterile pre-chilled mortar and pestle with 500 μL of sterile PBS and 50 mg sterile sand for friction [35].Contents were then centrifuged in a 1.5 mL vial at 9,500 g for 5 minutes.Supernatant was pipetted from the sand deposit, inserted into a new vial and stored at -20˚C.Genomic DNA was extracted from tick supernatant using TIA-Namp Genomic DNA Kit (Tiangen Biotech (Beijing) Co., LTD, Beijing, China) and tested for molecular detection of Rickettsia spp.targeting the citrate synthase gene (gltA) [36] and the outer-membrane protein gene (ompA) [37], as previously described (Table 1).For the molecular detection of Anaplasma spp.and Ehrlichia spp., the 16S rRNA gene [17] was targeted as previously described (Table 1).
Gel electrophoresis was used to evaluate amplified products using 1% (w/v) agarose gels stained with ethidium bromide at 120 V. Gels were analyzed using the Gel Doc EZ System (Bio-Rad, Hercules, California) with ultra-violet illumination.
A representative subset of positive amplicons were selected and directly sequenced using Sanger sequencing (Eton Biosciences, Inc., NC, USA).Sequencing results were then compared against the NCBI nucleotide database using the Standard Nucleotide BLAST application (http://www.ncbi.nlm.nih.gov/BLAST/) for species identification.Rickettsia spp.gltA and ompA sequences were used as confirmation of amplified Rickettsia spp.samples and Anaplasma/Ehrlichia spp.16S rRNA sequences were used as confirmation of amplified Anaplasma spp.Anaplasma/Ehrlichia spp., and Rickettsia spp.sequences were structured for phylogenetic relatedness using Molecular Evolutionary Genetics Analysis (MEGA) software, version 7.0.

Data analysis
Engorged tick infection status was compared to corresponding oviposited egg infection status for PCR-positive Rickettsia spp.and Anaplasma/Ehrlichia spp.samples, as well as sequence data.Transovarial transmission was considered to have occurred when the corresponding female tick and egg mass were both PCR positive.
Statistical analyses, including two-way frequencies with measures of association, were conducted using STATA 14.1 (StataCorp, College Station, TX).

Results
A total of 656 ticks were collected from 15 different locations across nine soums in five aimags.All ticks were morphologically identified as Dermacentor spp.Due to the size of larval and nymphal ticks collected, and the variety of Dermacentor spp.found in Mongolia, early life stage ticks were only identified to genus.All adult-fed ticks were morphologically identified to be D. nuttalli.Of the early life stage ticks collected from small mammals, 546 (88%) were larvae, 72 (12%) were nymphs.There were 588 (95%) of 618 ticks that were allocated into 42 larval and 18 nymphal pools (60 pools total).A total of 38 adult fed female ticks were collected from sheep and cattle.Of the 38 adult ticks collected, 33 laid eggs.

Molecular results
All  The evolutionary history was inferred using the Neighbor-Joining method [38].The bootstrap consensus tree inferred from 10,000 replicates is taken to represent the evolutionary history of the taxa analyzed [39].Branches corresponding to partitions reproduced in less than 50% bootstrap replicates are collapsed.The percentage of replicate trees in which the associated taxa clustered together in the bootstrap test (10,000 replicates) are shown next to the branches [39].The evolutionary distances were computed using the Kimura 2-parameter method and are in the units of the number of base substitutions per site [40].The analysis involved 17 nucleotide sequences.Codon positions included were 1st+2nd+3rd+Noncoding.All ambiguous positions were removed for each sequence pair.There were a total of 642 positions in the final dataset.Evolutionary analyses were conducted in MEGA7 [41].The evolutionary history was inferred using the Neighbor-Joining method [38].The bootstrap consensus tree inferred from 10,000 replicates is taken to represent the evolutionary history of the taxa analyzed [39].Branches corresponding to partitions reproduced in less than 50% bootstrap replicates are collapsed.The percentage of replicate trees in which the associated taxa clustered together in the bootstrap test (10,000 replicates) are shown next to the branches [39].The evolutionary distances were computed using the Kimura 2-parameter method and are in the units of the number of base substitutions per site [40].The analysis involved 18 nucleotide sequences.Codon positions included were 1st+2nd+3rd+Noncoding.All ambiguous positions were removed for each sequence pair.There were a total of 1254 positions in the final dataset.Evolutionary analyses were conducted in MEGA7 [41].
https://doi.org/10.1371/journal.pntd.0006696.g005The evolutionary history was inferred using the Neighbor-Joining method [38].The bootstrap consensus tree inferred from 10,000 replicates is taken to represent the evolutionary history of the taxa analyzed [39].Branches corresponding to partitions reproduced in less than 50% bootstrap replicates are collapsed.The percentage of replicate trees in which the associated taxa clustered together in the bootstrap test (10,000 replicates) are shown next to the branches [39].The evolutionary distances were computed using the Kimura 2-parameter method and are in the units of the number of base substitutions per site [40].The analysis involved 19 nucleotide sequences.Codon positions included were 1st+2nd+3rd+Noncoding.All ambiguous positions were removed for each sequence pair.There were a total of 1518 positions in the final dataset.Evolutionary analyses were conducted in MEGA7 [41]. https://doi.org/10.1371/journal.pntd.0006696.g006

Discussion
Few studies have evaluated transovarial transmission of tick-borne rickettsiae in Mongolia [42,43].As a result, data regarding transovarial transmission of Rickettsia spp.[44] are particularly sparse and it remains unclear what role transovarial transmission of Rickettsia spp.and Anaplasma/Ehrlichia spp.plays in the maintenance of tick-borne rickettsiae in Mongolia.Additionally, few studies have assessed tick-borne rickettsiae in tick larvae collected from small mammals [45,46].
It is often difficult to compare field surveys on tick-borne rickettsiae, because of varying sampling methods and sample sizes.However, the detection of R. raoultii found in this study show similarity with other studies conducted in Mongolia with 100% of Dermacentor spp.tick pools testing positive for R. raoultii [20,22].The positive molecular status of the majority of adult ticks were R. raoultii with few other ticks carrying uncharacterized rickettsiae, suggesting that the prevalence of R. raoultii in the areas in which the ticks were collected is quite high.
Though this study does not report animal host infection status, it can prove advantageous to test larvae collected from animal hosts as a method of determining if rickettsiae infect ticks by transovarial or transstadial transmission with paired animal health information.A positive infection status in ticks could occur by horizontal transmission of rickettsiae from an infected animal host to the tick.The alternative explanation is that the larvae found on the small mammals may be infected with the Rickettsia spp.by transovarial transmission.Transovarial transmission of Rickettsia spp.has been demonstrated in controlled laboratory settings and observed in nature.Laboratory demonstration of transovarial transmission has often included controlled concentration of the rickettsiae, artificial climactic conditions (temperature, humidity, etc.), and use of experimental hosts such as rabbits [44], mice, guinea pigs or capillary feeding [47].Among these laboratory-based transovarial studies, the transmission rate of Rickettsia spp.from mother to progeny has been shown to occur up to 100% across various tick species, including Dermacentor genus [44,48].Additionally, many laboratory-based transovarial studies have demonstrated the efficiency of transovarial transmission of Rickettsia spp.over multiple generations of ticks [44,49].Among studies observing naturally infected ticks transmitting Rickettsia spp. to progeny, prevalence has ranged from 30% to 100% [48].Though laboratory challenge studies determine the capability of transovarial transmission, observational studies of transovarial events can provide a risk assessment for transovarial transmission of tick-borne rickettsiae in a given region.
It is well documented that naturally occurring Rickettsia spp.are sustained through both transovarial and transstadial transmission in Dermacentor spp.ticks based on previous epidemiological research of Rickettsia spp. in Inner Mongolia, China, suggesting that there may be similar maintenance of Rickettsia spp. in Mongolia [50].Previous challenge studies have demonstrated transovarial transmission of R. raoultii in D. nuttalli, D. silvarum, D. marginatus, and D. reticulatus ticks.Transovarial transmission ranged from 43% to 100% prevalence depending on generation of tick infection, species of tick, and strain of R. raoultii.In comparison to the previous laboratory-based research reporting 43% to 99.5% of transovarial transmission of R. raoultii in D. nuttalli ticks [49], this study reports comparable transovarial transmission prevalence between adult and pooled eggs in D. nuttalli ticks at 91%.Though there has been no conclusive data reporting vertebrate hosts as a reservoir for R. raoultii, it has been suggested that D. marginatus and D. reticulatus serve as both vector and reservoir [51].Additionally, D. nuttalli ticks have been implicated as the primary vector for R. raoultii in Mongolia [20].
Our study's high prevalence of infected egg clutches suggesting transovarial transmission in conjunction with the substantial number of infected larvae found on small mammals, suggests that Dermacentor species (potentially D. nuttalli) ticks may serve not only as the primary vector, but also as the primary reservoir for R. raoultii in the northern region of Mongolia.
Though this study did not identify Anaplasma spp. or Ehrlichia spp. in larvae or nymphs, there was detection of an Anaplasma spp.most similar to A. ovis in two D. nuttalli fed adult ticks and in one tick egg pool.There have been reports of A. phagocytophilum transmitted transovarially with prevalence's ranging from 10% to 40% [52], however transovarial transmission of most Anaplasma spp.and Ehrlichia spp.are thought to occur at low frequencies or not at all [53,54].There has been research evaluating the role of transovarial transmission of Anaplasma spp.and Ehrlichia spp. with little success [55].To our knowledge, this may be the first documented report suggesting potential transovarial transmission of A. ovis in D. nuttalli ticks [56].There has been reports of this particular Anaplasma spp. in Dermacentor spp.ticks [57] being associated with history of infection in goats [11,12], sheep [11,14], cattle [11], and reindeer [16] throughout Mongolia and China.Though A. ovis is not known to cause human disease, the economic burden is great for individuals who rely on raising livestock for their income [58].Further research is needed to evaluate the efficiency and role of potential transovarial transmission of A. ovis.

Limitations
Like many tick pool studies, it is difficult to determine the exact prevalence of disease using this approach.Due to the nature of the maximum likelihood estimation calculation, the proportion of infected ticks with maximum likelihood of being Rickettsia spp.infected within tick pools cannot be calculated if 100% of sample pools are positive [59].Additionally, due to the pooling of tick eggs, this study was unable to determine a more precise proportion of transovarial transmission from an infected female tick to at least one progeny.Though data suggest that transovarial transmission for R. raoultii did occur, we were unable to determine how many progeny were infected.Additionally, by only screening infection status of egg mass, we are unable to discuss if infected larvae will hatch.Furthermore, Rickettsia spp.PCR primers have been shown to cross-react with Anaplasma spp.and Ehrlichia spp.However, this study also utilized a general screening assay for Anaplasma/Ehrlichia spp.and confirmation by sequencing, which allowed for greater confidence in the Rickettsia spp.PCR assay.

Conclusion
The indication that D. nuttalli ticks can serve as reservoirs for R. raoultii may warrants additional evaluation of transovarial and transstadial transmission of R. raoultii.Studies should focus on assessing tick eggs, either in smaller egg pools or individually, to determine the proportion of transovarial transmission as well as transstadial transmission for R. raoultii in eggs entering larval life stage, and larvae entering nymphal stages in the natural foci of Mongolia.Additionally, the testing of larvae from animal hosts and the environment should be further examined, preferably testing individual ticks instead of tick pools.Also, this report has identified a potentially novel transovarial transmission of A. ovis.Further investigation would be needed to determine the efficiency and prevalence of transovarial transmission of this rickettsiae.

S2 Dataset. Engorged adult tick data. (PDF)
Fig 2) from May 6 th to 22 nd , 2016.The second component of this study, examined larvae and nymphs removed from trapped small mammals across seven soums in three aimags situated in the Northern region of Mongolia from June 20 th to July 23 rd , 2016.Using a handheld global positioning system (GPS) unit (Juno Trimble Positions System, Sunnyvale, CA), latitudinal and longitudinal coordinates were collected at each small mammal collection site (Fig 3).

Fig 3 .
Fig 3. Map of larval/nymphal tick collection sites in Mongolia.GPS data points of tick collection sites were downloaded into ArcGIS 10.4 (ESRI, Redlands, CA).Maps were downloaded from Mongolian Environmental Health Geodatabase (http://www.eic.mn/).https://doi.org/10.1371/journal.pntd.0006696.g003 individual adult ticks and larval/nymphal tick pools were PCR-positive for Rickettsia spp.Subsequent PCR testing of paired eggs resulted in 91% (30/33) PCR positive among tick egg pools for Rickettsia spp.Sequencing data revealed Rickettsia spp.shared !99% identity with R. raoultii ompA (Accession numbers MH234455 and MH234456) shown in the phylogenetic analysis (Fig 4).A majority (23/32) of gltA sequences shared !99% identity with R. raoultii (Accession numbers MH208721 and MH208722) shown in the phylogenetic analysis (Fig 5), however 9/32 sequences were considered inconclusive, falling between 84%-95% identity with R. raoultii.Of the 38 engorged adult ticks collected, two ticks (5%) were PCR-positive for Anaplasma/ Ehrlichia spp., while none of the larval/nymphal pools were PCR-positive for Anaplasma/ Ehrlichia spp.Additionally, one pool of eggs laid by an Anaplasma/Ehrlichia spp.-positive engorged adult female tick, was also found to be PCR-positive for Anaplasma/Ehrlichia spp.All PCR-positive Anaplasma/Ehrlichia spp.ticks and the positive egg clutch were further examined using a sequencing approach to identify the infecting rickettsial species.Sequencing results indicated that the Anaplasma/Ehrlichia spp.positive egg clutch and corresponding

Fig 4 .
Fig 4. Evolutionary relationships of Rickettsia spp.ompA.The evolutionary history was inferred using the Neighbor-Joining method[38].The bootstrap consensus tree inferred from 10,000 replicates is taken to represent the evolutionary history of the taxa analyzed[39].Branches corresponding to partitions reproduced in less than 50% bootstrap replicates are collapsed.The percentage of replicate trees in which the associated taxa clustered together in the bootstrap test (10,000 replicates) are shown next to the branches[39].The evolutionary distances were computed using the Kimura 2-parameter method and are in the units of the number of base substitutions per site[40].The analysis involved 17 nucleotide sequences.Codon positions included were 1st+2nd+3rd+Noncoding.All ambiguous positions were removed for each sequence pair.There were a total of 642 positions in the final dataset.Evolutionary analyses were conducted in MEGA7[41].

Fig 5 .
Fig 5. Evolutionary relationships of Rickettsia spp.gltA.The evolutionary history was inferred using the Neighbor-Joining method[38].The bootstrap consensus tree inferred from 10,000 replicates is taken to represent the evolutionary history of the taxa analyzed[39].Branches corresponding to partitions reproduced in less than 50% bootstrap replicates are collapsed.The percentage of replicate trees in which the associated taxa clustered together in the bootstrap test (10,000 replicates) are shown next to the branches[39].The evolutionary distances were computed using the Kimura 2-parameter method and are in the units of the number of base substitutions per site[40].The analysis involved 18 nucleotide sequences.Codon positions included were 1st+2nd+3rd+Noncoding.All ambiguous positions were removed for each sequence pair.There were a total of 1254 positions in the final dataset.Evolutionary analyses were conducted in MEGA7[41].

Fig 6 .
Fig 6.Evolutionary relationships of Anaplasma spp.16S rRNA gene.The evolutionary history was inferred using the Neighbor-Joining method[38].The bootstrap consensus tree inferred from 10,000 replicates is taken to represent the evolutionary history of the taxa analyzed[39].Branches corresponding to partitions reproduced in less than 50% bootstrap replicates are collapsed.The percentage of replicate trees in which the associated taxa clustered together in the bootstrap test (10,000 replicates) are shown next to the branches[39].The evolutionary distances were computed using the Kimura 2-parameter method and are in the units of the number of base substitutions per site[40].The analysis involved 19 nucleotide sequences.Codon positions included were 1st+2nd+3rd+Noncoding.All ambiguous positions were removed for each sequence pair.There were a total of 1518 positions in the final dataset.Evolutionary analyses were conducted in MEGA7[41].