Not Only Leptotrombidium spp. an Annotated Checklist of Chigger Mites (Actinotrichida: Trombiculidae) Associated with Bacterial Pathogens

Mites of the family Trombiculidae are known for playing a role in maintaining and spreading the scrub typhus etiologic agent, an intracellular Gram-negative bacterium, Orientia tsutsugamushi. Species of the genus Leptotrombidium are investigated most thoroughly, particularly in SE Asia, and a few are proven vectors for the pathogen. The mentioned association, however, is not the only one among trombiculids. Here, we present a list of chiggers indicated in the literature as positive for bacterial pathogens, tested throughout almost 100 years of research. Taxonomic identities of trombiculids follow recent revisions and checklists. Results point at 100 species, from 28 genera, evidenced for association with 31 bacterial taxa. Pathogen-positive mites constitute around 3.3% of the total number of species comprising the family. Discussed arachnids inhabit six biogeographic realms and represent free-living instars as well as external and internal parasites of rodents, soricomorphs, scadents, lagomorphs, peramelemorphs, bats, passerine birds, reptiles and humans. A variety of so far detected bacteria, including novel species, along with the mites’ vast geographical distribution and parasitism on differentiated hosts, indicate that revealing of more cases of Trombiculidae-pathogens association is highly probable, especially utilizing the newest techniques enabling a large-scale bacterial communities survey.

Trombiculid larvae feed basically on dissolved connective tissue similar in composition to a plasma, however, in single cases, ingestion of blood has been also reported [12][13][14]. For the purpose of food intake, after reaching a parasitope, mites pierce the host's skin with Associated with O. tsutsugamushi. Parasites collected from T. glis in Thailand [24].

L. umbricola Nadchatram and Dohany, 1980
Host-questing larvae tested positive for O. tsutsugamushi in Malaysia. Collected with the method of black plates [83,127].
Remarks: unengorged larvae of L. vivericola Vercammen-Grandjean and Langston, 1976 found in Malaysia by Dohany et al. [128] were also reported as O. tsutsugamushi-associated, however, the subsequent revision of the material proved the first species determination wrong and larvae were then assigned to L. umbricola [127].
Remarks: According to Stekolnikov et al. [18], all records concerning N. autumnalis should be thoroughly verified after proper mounting on microscopic slides, due to considerable resemblance of this species to N. inopinata. Proven misidentification was reported from Turkey.

S. ligula Radford, 1946
Individuals captured on R. rattus and S. murinus tested positive for O. tsutsugamushi in India. The association was mentioned also from Malaya and Pakistan [14,49,140].
Mites of the genus are reported to carry O. tsutsugamushi and Rickettsia sp. Bacteriapositive larvae were obtained from R. tanezumi, B. savilei, and B. indica in Thailand [23,47].

W. micropelta (Traub and Evans, 1957)
Larvae of the species contained genetic material of O. tsutsugamushi. Collected from B. indica and M. surifer in Thailand [23].
The above records are graphically summarized in Figures 1 and 2.
The above records are graphically summarized in Figures 1 and 2.
Occurrence frequency of particular bacteria (genera and species) as well as bacterial families associated with Trombiculidae are illustrated in Figures 3 and 4.   again, Leptotrombidium spp. are positive for the highest quantity of pathogenic taxa (14), next is Neotrombicula (11) and Eutrombicula (7) (Figure 2). Rickettsiaceae is the dominant family detected in Trombiculidae. Orientia tsutsugamushi infection was confirmed in 79 cases, and is followed by R. typhi and R. felis (7 cases, each), to name only the most frequent rickettsial species. Considerable incidence also characterizes Borreliaceae-infection with Borrelia spp. was observed in 11 cases (Figures 3 and 4). Hereby discussed data concern only non-artificial infections of mites originating from the natural environment. We excluded experimentally induced associations, such as acquisition of B. garinii and C. burnetii by larvae of the European harvest mite after parasitizing infected rodents under laboratory conditions, or obtaining O. tsutsugamushi-positive larvae of the common Northamerican chigger resulting from experimental feeding upon host with rickettsaemia [42,54,133]. Associations with viruses, for example, between Hantaan virus and L. scutellare [147,148], however very interesting, were omitted too. Rickettsiales, the obligate intracellular parasites [66], are especially prevalent within chigger mites. With the highest probability, it is the result of the already described larval feeding mechanism (dissolving and ingesting cell contents from host's epidermis and dermis) [16], combined with the affinity of rickettsial bacteria to reside in the connective tissue, including its external layers forming the skin [149][150][151][152]. Similar tendency is characteristic for Borrelia spp. spirochetes [153][154][155]. Producing the stylostome during food intake is a common feature of trombiculids, however, its length, width and wall structure is differentiated among genera [16,156]. Hase et al. proposed three types of stylostomes depending on the longitude (epidermal-the shortest, mesenchymal-the longest and a mix of the two) based on observations of Leptotrombidium spp feeding. Authors hypothesized that the canal structure might have been related to the mite's ability to acquire bacteria, as larvae of L. intermedium (the only species to form epidermal stylostome) turned out to be pathogen-free [157]. This assumption was subsequently refuted as later studies revealed the association between O. tsutsugamushi and L. intermedium. Moreover, Neotrombicula pomeranzevi, Miyatrombicula esoensis (Sasa and Ogata, 1953) and Kepkatrombicula desaleri (Methlagl, 1928) produce the longest feeding canals, thus penetrating host tissues relatively deep, albeit the ability to ingest pathogens by the two latter species still awaits confirmation in laboratory tests. On the other hand, H. zachvatkini and Leptotrombidium spp. Create shorter canals which widen in time and were reported as bacteria-positive quite frequently (Figures 1 and 2). Cheladonta costulata, known for parasitism entailing submergence of almost the entire body into host tissues and producing feeding tubes of very variable length, was capable of acquiring rickettsiae too [16,156]. Considering the fact that the current number of pathogen-associated trombiculids is 100, whereas formation of the stylostome has been examined in a fracture of this group, the relation between the structure of the feeding canal and acquisition of bacteria is still to be verified in the research involving more species. A common feature of the remaining bacterial genera detected in trombiculids (i.e., Bartonella, Coxiella, Francisella, Leptospira, Mycobacterium) is their natural presence in organisms of Rodentia and Soricomorpha [42,[158][159][160][161][162], which are not only preferable hosts of chiggers in general but also were sources from which infected parasitic instars were collected. Named microorganisms, however, (unlike Rickettsiaceae and Borreliaceae) do not demonstrate high affinity to external layers of the connective tissue, but are reported to reside mostly in phagocytes, endothelium, erythrocytes and kidney cells as well as in soil and water [163][164][165][166][167][168]. This may potentially explain lower incidence of these genera in mites (Figures 3 and 4). Inasmuch some bacterial species from the above genera can also dwell in the moist microhabitats (also preferred by many trombiculids), it cannot be excluded that their presence in mites might be a result of contamination.
A term association between mites and pathogens has been deliberately applied as vectorship, i.e., capability of effectively transmitting bacteria to humans (or other vertebrates) is not proven for all pathogen-positive chigger species and the presence of microorganisms may be resulting from ingesting dissolved host tissues (i.e., pathogen's reservoir), especially when engorged larvae are preserved and tested shortly after being detached from the host.
Confirmed vectors of Orientia spp., e.g., L. deliense, L. akamushi, L. scutellare or H. antarctica, meet the criteria formulated by Traub and Wisseman: natural infection with a pathogen, ability to infect a host (by feeding process, after being crushed on host's body or with infected feaces), high prevalence in a given area and, tendency for parasitizing humans. The latter point regards of course only diseases plaguing people and it is not essential for the mechanism of vectorship as such. As Trombiculidae, unlike, e.g., Ixodidae or Macronyssidae, are parasitic once in a lifetime, to successfully transmit bacteria larvae have to acquire them via transstadial and transovarial transmission from the parental generation [14,54]. The latter two phenomena also have been analyzed mainly in Leptotrombidium spp. so far e.g., [19][20][21]. At the same time, observations of bacteria-positive, unengorged larvae from Eutrombicula, Herpetacarus, Microtrombicula, Miyatrombicula, Neotrombicula, Odontocarus and Walchia genera along with the infected deutonymphs and successful experimental infections, indicate that the possibility of effective pathogen transmission remains high in a variety of chigger mite species.
One should bear in mind, yet, that detailed research on vector competence of the particular mites species is often hindered by two issues. The first one is extremely varied numbers of mites from particular species in collected samples-ranging from thousands of individuals of the most common taxa, to single ones of the most infrequent (H.M. personal observation). This is illustrated by results of chigger collection in NW Russia wherein thousands of H. zachvatkini larvae were present in contrast to nine of A. latyshevi (Schluger, 1955) and one N. absoluta Schluger, 1966 larva. Significant disproportions in field-collected chiggers were also recorded during a survey performed in India-the most common species L. deliense and L. insigne  [92,169,170]. Although single specimens can be easily tested for the presence of bacteria, results based on small samples are not fully representative. Moreover, verification of transstadial and transovarial transmissions would require experimental rearing of a few mite generations based on a considerable number of chiggers, with consideration of their mortality [171]. The second problem is the correct identification of potential vector species. An example of this kind of impediment is already signalized misidentification between closely related N. autumnalis (considered the most common European chigger, however, absent in some countries (H.M. personal observation) and N. inopinata (species often determined after the proper identification of specimens previously assigned to 'N. autumnalis') [18]. Furthermore, Ponnusamy et al. have recently reported on problems with matching COI sequences of Rickettsia-positive Eutrombicula sp. and Leptotrombidium sp., previously determined upon morphological criteria, with corresponding data in the GeneBank using the BLAST tool [64]. This problem is not limited to chiggers inhabiting Nearctic, as the deficiency of reference sequences in the GeneBank is still noticeable (such data are available for c. 80 nominal species only). At the same time, obtaining COI sequences of L. imphalum and L. chaigraiensis led to separation of these species, formerly considered as one [86,88].
Identification difficulties should not suppress the research on pathogen transmission by Trombiculidae inasmuch the most accurate species determination can be achieved by simultaneous application of morphological and molecular tools, as it has been already implemented in some studies [88,[172][173][174]. Moreover, this issue necessitates further exploration as bacterial diseases potentially transmitted by chiggers are largely negligent in parts of the globe where they are apparently absent, as it was emphasized by Weitzel et al. [39]. For example, not only is the scrub typhus present in the subantarctic Chile, but also its etiological agent turned out to be a novel Orientia species-Candidatus Orientia chiloensis. The pathogen was effectively transmitted to humans by H. antarctica [36,39,73,175]. Ana-logical instances of bush typhus bacterium, distinct from O. tsutsugamushi (i.e., Candidatus Orientia chuto), detected in Trombiculidae occurring outside the 'tsutsugamushi triangle' come from Kenya and Saudi Arabia. From the latter country, O. chuto-positive patient was reported as well [37,38,130,143]. Other proposed bacterial species (i.e., Candidatus rickettsia colombianesi, Ca. rickettsia jinxinensis and Ca. rickettsia leptotrombidium) harbored by Trombiculidae, along with the most recent and the first ever findings of rickettsiae in chiggers from North Carolina (USA) [64] only reaffirm the sense of continual survey on, otherwise unrecognized and hidden, pathogens and their transmission routes. Exploring the issue is also justified by non-decreasing importance of parasitic and bacterial zoonoses in general [42] and can be greatly supported by the highly effective technologies (e.g., RAPD-PCR fingerprinting or high throughput sequencing (HTS) related techniques such as DNA metabarcoding), application of which can result in further contributions to the hereby reviewed matter.

Materials and Methods
Google scholar, PubMed and Scopus databases were searched with bacteria, chigger mites, detection, pathogen, spirochaetes, Trombiculidae, trombiculid terms. Collected records cover almost 100 years (from 1924 to 2022) of the research on the association between chiggers and bacterial pathogens, detected by means of microscopic, culturing, serological and molecular examinations performed worldwide (Section 2.1). Testing techniques as well as the very history of studies on Trombiculidae-bacteria relation are summarized and thoroughly described in the literature [10,32].
Taxonomic nomenclature and systematics of Trombiculidae follow elaborations, taking into account morphological and molecular data [2,[86][87][88]103]. Species authorities, scientific names as well as common names and systematics of vertebrates are given at first mention, according to Wilson and Reeder and IUCN Red List [176,177].
Species of Trombiculidae lacking authority and not present in the above reference sources are placed in a separate subsection (2.2) but are not included in plots and calculations.

Conclusions
The share of pathogen-associated trombiculids is low in comparison with the total number of nominal species comprising the Trombiculidae family. Nonetheless, so far revealed bacteria-positive mites are characterized by harboring differentiated bacterial species, vast geographical distribution and association with a variety of hosts. Moreover, the present summary also points at cases of relatively recently discovered novel bacterial species and localities wherein the discussed microorganisms were apparently absent to date. This knowledge, combined with the unwavering significance of zoonotic bacterioses and the recognized mechanisms of pathogens circulation in chigger populations, are premises that the actual number of Trombiculidae-bacteria associations is not limited to the cases presented. An assumption can be made that the continual microbiological testing of chiggers, especially when supported with the fast and highly effective technologies, will result in further findings.