Eco-epidemiology of rodent-associated trombiculid mites and infection with Orientia spp. in Southern Chile

Background Scrub typhus is a potentially severe infection caused by bacteria of the genus Orientia, endemic in Asia-Pacific and recently discovered in southern Chile. The presented study aimed to determine the prevalence and species richness of rodent-associated trombiculid mites and their infection with Orientia spp. in different areas of two regions in southern Chile. Methodology/Principal findings During summer 2020, trombiculid mites were collected from rodents captured in three areas in southern Chile known to be endemic for scrub typhus (Cochamó and Chiloé Island in the Los Lagos Region and Tortel in the Aysén Region). A total of 132 rodents belonging to five species were captured using Sherman-like traps; 89.4% were infested with trombiculids. Mite specimens were morphologically identified and subsequently tested by Orientia-specific qPCR. Six mite species were identified. Among chigger-infested rodents, 33.9% carried Orientia-positive mites; this rate was higher in Tortel (63.8%) than in Cochamó (45.0%) and Chiloé Island (2.0%). The analysis of individual mites (n = 901) revealed that 31.2% of Herpetacarus antarctica samples (n = 202) were positive for Orientia DNA; the prevalence was 7.0% in Paratrombicula neuquenensis (n = 213), 6.9% in Herpetacarus eloisae (n = 144), 3.6% in Argentinacarus expansus (n = 55), and 0% in Paratrombicula goffi (n = 110) and Quadraseta chiloensis (n = 177). The southernmost site (Tortel) showed the highest rates of trombiculid infestation, trombiculid load, and Orientia infection in the captured rodents. Conclusions/Significance Our study provides new insights into the trombiculid fauna and prevalence of Orientia in mites collected from wild rodents in southern Chile. Orientia DNA was detected in four of the six mite species. Rates of infestation, mite loads, and Orientia prevalences differed geographically and were highest in the Aysén Region. Our data improve our knowledge on possible vectors of scrub typhus and their distribution in Chile.


Introduction
Scrub typhus threatens over one billion people in the Asia-Pacific region, where, despite its high morbidity and mortality, it is considered a neglected disease [1]. Initially, scrub typhus was thought to be restricted to an area known as "tsutsugamushi triangle"; however, the clinical cases reported from South America and the Middle East in the last 15 years, together with serological evidence from Africa and Latin America, suggest a much wider distribution of the disease [2]. Recent data from Chile showed that endemic areas included various geographical and climatic regions over a distance of almost 2,000 km [3][4][5].
Scrub typhus is caused by obligate intracellular bacteria of the genus Orientia, belonging to the Rickettsiacea family [6], with three species described so far, Orientia tsutsugamushi occurring in Asia-Pacific, Candidatus Orientia chuto (from United Arab Emirates), and Candidatus Orientia chiloensis (from Chile) [7,8]. O. tsutsugamushi is transmitted by mites of the family Trombiculidae (Acari: Trombidiformes), mostly belonging to the genus Leptotrombidium, which, through transovarial and transstadial transmission, also serve as the pathogen's reservoir [9]. The mites' ectoparasitic larvae, called chiggers, have low host specificity [10], but rodents seem to be the main determinants for the maintenance of stable mite populations in Asia-Pacific [11,12]. In Chile, the knowledge of the eco-epidemiological aspects of scrub typhus and its vectors is incomplete. Up to now, 25 trombiculid mite species of 12 genera have been recorded in Chile, mainly in relation to reptiles [13][14][15][16]. A first investigation of rodentassociated mites from scrub typhus endemic localities on Chiloé Island reported three chigger species, of which two were new to science; one novel species, Herpetacarus eloisae, was found positive for Orientia DNA, suggesting its capacity as vector and reservoir of scrub typhus [16,17]. In addition, a recent report proved Herpetacarus antarctica as vector of scrub typhus in the Aysén Region in southern Chile [18].
The here presented study analyzed the rodent-associated trombiculid fauna and the possible role of trombiculids as vectors and reservoirs of Orientia spp. in three areas with endemic scrub typhus in southern Chile.

Ethics statement
The study was approved by the Scientific Ethics Committee for the Care of Animals and the Environment of Pontificia Universidad Católica de Chile (N˚160816007, January 28, 2020) and by the Agricultural and Livestock Service (SAG) (Exempt Resolution No.: 858/2020, February 3, 2020).

Study sites
The study was carried out in three areas in southern Chile, Cochamó (41˚45'40"S, 72˚5'35"W), Chiloé Island (41˚52'15"S, 73˚48'58"W), both in the Los Lagos Region, and Tortel (47˚47'54"S, 73˚32'24"W) located in the Aysén Region (Fig 1). In all three study areas cases of scrub typhus have been detected by our research group [4,19]. Cochamó has a temperate oceanic climate, with an annual rainfall of 3,407mm and an average annual temperature in the Andean valleys of 6.6˚C; however, in summer, average temperature raises up to 13.2˚C and rainfall decreases [20]. In Cochamó the two capture sites (S1 and S2) are covered by Valdivian temperate rainforest with a certain degree of human intervention such as thinning and construction sites [20]. Chiloé Island also has temperate oceanic climate with an annual rainfall of 2,090 mm and average annual temperature of 12˚C [21]. The landscape consists of fragments of Valdivian temperate rainforest of different sizes, shapes, degrees of isolation and degradation. The selected capture sites (S3 and S4) correspond to localities N˚3 and N˚4, respectively, of our previous study [17]. Tortel has a subpolar oceanic climate (cold temperate, rainy, coastal), annual rainfall of 3,500-4,000 mm, and average temperature of 6˚C-8˚C. In this area, sampling was carried out at two sites, Aumén (S5) and Laguna Caiquenes (S6), about 18km southeast of the village Caleta Tortel. The vegetation consists of Valdivian temperate rainforest with bushy thicket with a lower floor of hardy perennial grassland [22].

Rodent trapping
Rodent sampling was carried out during end of February and beginning of March 2020, within the austral summer period. In each study area, two sampling sites were surveyed. At each site, between 75-80 Sherman-type traps (300 × 100 × 110 mm) were activated at 5 meters distance, placed under bushes or fallen logs or in burrows. Traps were placed simultaneously in each site during 5 consecutive nights with an effort of 375-400 trap-nights per site, 750-800 trapnights per area, and total of 2,350 trap-nights. Captured rodents were transferred to a central processing tent, where they were chemically immobilized in an induction chamber containing cotton soaked with isoflurane (USP, Baxter; 1 mL of isoflurane per 500 mL of chamber volume) [17]. After sedation, male and juvenile rodents (both sexes) were euthanized by cervical dislocation [23], while adult females were released at the capture site after examination. Each rodent was measured with a caliper (Serie 500, Mitutoyo, Japan), weighed (Spring Scale 10100, Pesola, Switzerland), and morphologically identified, as previously described [24]. Subsequently, the fur of each rodent was inspected using fine-tipped tweezers to collect trombiculid mites, which were mostly located in the ears (Fig 2). Trombiculids from individual rodents were stored in vials with 96% ethanol for subsequent taxonomic and molecular analyses. Rodent's tissues (i.e. liver, spleen and lungs) were also obtained and deposited in 96% ethanol for further molecular studies.
Due to the study team's exposure risk to Andes virus, causing hantavirus cardiopulmonary syndrome [25], rodents were handled following biosafety recommendations of the CDC and the American Society of Mammalogists [26]. Thus, researchers were equipped with a full-face respirators (6000 Series, 3M, Minnesota, USA) equipped with P100 ultrafine particle filters (2000 Series, 3M), disposable coveralls, shoe covers, and latex gloves. Traps were disinfected daily with hypochlorite solution. In addition, members of the fieldwork team were advised and followed-up for 5 weeks after the field period to recognize possible signs and symptoms of scrub typhus [27].

Mite identification
The taxonomic analyses of mite specimens were carried out at the Instituto de Medicina Preventiva Veterinaria, Universidad Austral de Chile in Valdivia, Chile. Initially, mites were

PLOS NEGLECTED TROPICAL DISEASES
observed under a stereo microscope (SZ61 Olympus, Tokyo, Japan), which permitted to separate different morphotypes. Approximately 20 samples of each morphotype were rinsed with Nesbitt solution and individually mounted with Berlese medium between a microscope slide and coverslip [28]. Mounted specimens were further analyzed under an optical microscope (BS-2030T, BestScope, Beijing, China) with a magnification of 640×, following keys proposed by Brennan and Goff [29] as well as Stekolnikov and González-Acuña [13]. The remaining samples were examined under a fluorescence optical microscope (BS-2030FT, Bestscope), which allowed to observe species-specific morphological criteria, e.g. shape of the scutum and its setae (Fig 3), as previously described [30]. This permitted to separate and store individual and pooled mites of known species for further molecular analyses. A subset of mounted samples of each mite species was sent to the Zoological Institute of the Russian Academy of Sciences in Saint Petersburg, Russia, to confirm the morphological identification.

Molecular detection of Orientia DNA in trombiculid mites
With the exception of mounted specimens, a subset of mites was tested for the presence of Orientia DNA as individuals or as pools. Of each chigger species (collected from an individual rodent) approximately 10 samples were tested as individual mites, while the remaining specimens were analyzed as pools of 3-60 mites in Cochamó and Chiloé Island and of 10-100 mites in Tortel. Individual and pooled samples were dried at 72˚C for 3 hours in an oven to eliminate excess ethanol and stored at -80˚C in a mixture of 180 μL of lysis buffer and 20 μL of proteinase K (Qiagen, Hilden, Germany). To mechanically disrupt the mites' exoskeletons, we used a thermal method, based on previously described protocols [31,32]. Samples were thawed at room temperature and then subjected to three freeze-thaw cycles (2 minutes in liquid nitrogen and 2 minutes at 70˚C), followed by incubation at 56˚C for 3 hours in a thermal bath. After this procedure, DNA was extracted using the manufacturer's instructions for the QIAamp DNA Mini Kit (Qiagen). Finally, samples were eluted in 50 μL of buffer AE (Qiagen) and stored at 4˚C, if molecular testing was carried out within 24 hours, or at -80˚C, if this was done later. For the molecular detection of Orientia DNA, real-time quantitative polymerase chain reaction (qPCR) targeting the rrs gene (16S RNA) was used, which has recently been developed to detect Orientia species including strains from Chile [33].

Data analysis
Firstly, we analyzed the pattern of trombiculid infestation on the rodent species captured in different sites, which included infestation rates, mean abundance per host (i.e. chiggers index), and mean species richness per host. Secondly, we assessed whether the number of chigger species and the total load of chiggers parasitizing each host differed between areas using Kruskal Wallis test and Dunn's Test for post-hoc comparisons. Finally, we performed chi-square test and Bonferroni adjustment for post-hoc analyses to compare Orientia-positive mite rates in wild rodents between host species and sampling areas. All statistical analyses were performed using R software [34].

Discussion
Trombiculid mites are important but neglected vectors of rickettsial infections and other pathogens [36]. In the Asia-Pacific region, Leptotrombidium species serve as vectors of O. tsutsugamushi [9]. First data from Sub-Saharan Africa demonstrated that rodent-associated chiggers in Kenya, belonging to the genera Neotrombicula and Microtrombicula, contained DNA sequences closely resembling those of Ca. O. chuto [37]; in this region, however, clinical cases of scrub typhus have yet not been reported. In Chile, the knowledge of the trombiculid mite fauna and its possible role in the transmission of the recently discovered third Orientia species (Ca. O. chiloensis) is scarce. A first field study on Chiloé Island in 2018 found one known and two new rodent-associated chigger species; one of the novel species (H. eloisae) was infected with Orientia [17]. Finally, a recent report described chiggers collected from humans and environment in Tortel, providing first evidence that H. antarctica serves as vector and reservoir of scrub typhus [18].

PLOS NEGLECTED TROPICAL DISEASES
Most information on the ecology of the vectors of scrub typhus derives from Asia-Pacific [12]. In this region, the species richness of trombiculids is higher in areas of mid-latitudes and decreases in higher latitudes [10,38,39]. The present study suggests a similar gradient in Chile, as a total of five species of trombiculids were found in the Los Lagos Region, while only a single species was identified in the more southern Aysén Region. Chigger infestation rates of different wild rodent species detected in our study ranged from 57.1% to 100%, which is compatible to reports in Asia-Pacific [12]. As reported from other regions, we saw no host specificity, as larval mites feed opportunistically on a variety of hosts [40][41][42]. A recent systematic review of the ecology of scrub typhus by Elliot et al. [12], found that 16% of studies reported chigger index values of infested rodents. Our results from Cochamó and Chiloé Island are within the range of these studies; however, the values detected on L. micropus in the two sites of Tortel (i.e. 251.1 and 175.6 mites/rodent) are among the highest worldwide [12].
In Asia-Pacific the spatial distribution of chiggers is heterogeneous, with distinct areas of high mite density, so called "mite islands" [12,43]. According to Chakraborty and Sarma [44], various environments including sandy beaches, mountains, rainforests, riverbanks, and grass lands might provide optimal conditions for mite proliferation. The resulting higher prevalence and richness of chigger mites have been associated to seasonal scrub typhus outbreaks [45]. Such favorable conditions for chigger proliferation might also be present in our study areas, which all represented sites, where clinical scrub typhus cases had been recorded in previous years. The high infestation rates and chigger index values in some of the study sites support the concept of mite islands within the Chilean chigger ecosystems; however, future ecological studies are needed to assess whether a combination of factors such as extreme temperatures in summer months could be favoring the life cycle of chiggers and the bacteria over a wider spatial range around possible hot spots. Apart from the noted differences in mite fauna and density, the infection rates with Orientia varied significantly. The mite abundance and Orientia infection rate was especially striking in Tortel. As mentioned above, our field researchers recorded multiple mite bites and one suffered from scrub typhus; unengorged mites were also ubiquitous noted on lower vegetation [18]. In addition, the Orientia infection rate of individual mites in this area was 24-times and 100-times higher than in Cochamó and Chiloé Island (Table 3), respectively. Our findings in Chiloé Island might have been influenced by temporal aspects or sampling error, such as changes in temperature and humidity among years and reduced trapping effort, which might be responsible for the lower captured rodents and Orientia infection in mites between the here reported results and a previous field project in 2018 at similar localities [17]. Similar annual variations of mite species have been described in the Asia-Pacific [46][47][48]. Further studies are necessary to understand these fluctuations, which could also be caused by variations of the rodent population or anthropogenic changes in the habitat.
According to a review from 2019, about 50 of the >3000 known species of Trombiculidae worldwide are known to be positive to Orientia and; among those, approximately 15 species, all belonging to the genus Leptotrombidium are confirmed or possible vectors of scrub typhus [12]. The cited review does not include recent data from Chile, according to which Orientia DNA was found in a novel trombiculid species, H. eloisae, parasitizing rodents collected on Chiloé Island, while two other mite species, P. goffi and Q. chiloensis, tested negative [16,17]. The present study confirmed this; among the four species collected in this area, only H. eloisae carried Orientia DNA. The report from 2022 described Orientia DNA in H. antarctica samples, which were collected from a member of the research team in Tortel, who later developed scrub typhus, thus providing first direct evidence of the scrub typhus transmission by a trombiculid mite not belonging to the genus Leptotrombidium [18]. The here presented study detected the highest Orientia infection rate in this species. As H. antarctica has been described in the Magallanes Region in southern Chile [13], it is probably the vector in Tierra del Fuego in the extreme south of Chile, where scrub typhus cases were identified since 2020 [5]. In Cochamó, Los Lagos Region, which has not been sampled before, we collected H. eloisae, P. neuquenensis, Q. chiloensis, and A. expansus. Of 213 P. neuquenensis samples, 15 (7.0%) were Orientia positive; however, seven of those were collected from rodents co-infested with H. eloisae, which were Orientia-positive. Also two of 43 (4.7%) A. expansus specimens contained Orientia DNA, both originated from rodents co-infested with Orientia-positive H. eloisae. This new finding could indicate a broader spectrum of vector species of Orientia spp. in Chile. However, the data have to be interpreted cautiously and require further studies. From Asia, it has been suggested that Orientia infection may be transmitted passively between co-feeding chiggers [12,43,49]. This phenomenon, which has also been described for Rickettsia spp. in other arthropods [50], might lead to the detection of Orientia in mite species that do not serve as natural vector/reservoir. For O. tsutsugamushi, this has also been proven in animal experiments as well as suggested in field studies from Thailand [51,52]. In Cochamó and Chiloé Island, co-infestation with more than one trombiculid species occurred in >60% of rodents (Table 2). Another possible explanation could be the ingestion of Orientia DNA through tissue fluids of infected rodents, since in a previous study tissue samples of 34% of rodents from Chiloé Island tested positive for Orientia DNA [53]. Orientia infection rates among chigger specimens from Cochamó and Chiloé Island (i.e. 0.3%-7.4%) were similar to those reported in Asia-Pacific (from 0.6%-5%); the rate in Tortel (31.2%), however, was six times higher than the maximum value reported in a recent review [12]. As reported recently, infection rates in samples collected from the environment (6/18, 33.3%) and from humans (10/22, 45.4%) were similarly high [18], indicating that Tortel represents a "hot spot" area for scrub typhus.
The present investigation improves our knowledge on the eco-epidemiology of Orientia and its vectors in southern Chile, providing new data on the rodent-associated mite fauna and abundance in different scrub typhus endemic regions. Two species of Herpetacarus, H. antarctica and H. eloisae, were infected with Orientia spp. in all three study areas, which further strengthens their probable role as vectors of scrub typhus. If other mite species, which were found to contain Orientia DNA in Cochamó, might also have vector capacity, requires further studies and confirmation. The high density of Orientia-infected chiggers in the most southernmost study site highlights that in certain localities there is a high risk to acquire scrub typhus. Outdoor activities such as camping, hiking, fishing, and mountaineering or ecological excursions and wildlife photography are common in southern Chile and associated to scrub typhus cases [19]. The same applies for daily activities or work with contact to vegetation or firewood in rural areas in southern Chile [27]. Therefore, eco-epidemiological studies are not only important to understand the lifecycle of the new Chilean Orientia species and its vectors, but also to provide advice for health authorities to prevent and manage cases of scrub typhus in the endemic regions.