Molecular detection of Rickettsia species and host associations of Laelaps mites (Acari: Laelapidae) in Taiwan.

Various rickettsiae have recently been detected in Laelaps mites (Acari: Laelapidae), which are common ectoparasites of rodents; however, investigations on this topic remain very scarce, particularly in Asia. In the present study, shrews and rodents were trapped from 2006 to 2010 in eight lowland regions of Taiwan (< 500 m in elevation) to collect associated Laelaps mites, from which Rickettsia-a group of emerging pathogens-were detected and identified by assaying the gltA and ompB genes. A total of 853 Laelaps mites of at least four species were collected from a sample of 1004 small mammals that included one shrew and 10 rodent species. Rattus losea was the most common species (44.9% of total hosts) and hosted the highest percentage of mites (76.6% of total mites). Laelaps nuttalli was the most abundant mite species (51.7% of total mites), followed by Laelaps echidninus (24.2%), Laelaps sedlaceki (23.1%), and Laelaps myonyssognathus (0.2%). Notably, Rickettsia species with the highest similarity to spotted fever group (SFG) rickettsiae were identified from seven of the 72 pools of Laelaps mites. The presence of SFG rickettsiae in hematophagous Laelaps mites, particularly including species that are closely associated with commensal rodents in frequent contact with humans, calls for further investigation on the competence of Laelaps mites in transmitting rickettsiae.

Introduction

Small mammals, including rodents and shrews, are typically infested with a variety of ectoparasites, including fleas, lice, mites, and ticks. Some of these ectoparasites are capable of transmitting causative agents of diseases to humans, with these being more thoroughly studied by researchers. Most prominent among these are hard ticks (Acari: Ixodidae), which can transmit human diseases such as anaplasmosis, babesiosis, Lyme borreliosis, spotted fever, and tick-borne encephalitis (Dantas-Torres et al. 2012). Fleas also receive great attention for their central role as vectors of Bartonella, flea-borne spotted fever, murine typhus, and plague (Bitam et al. 2010). In comparison, the potential role of mites (Acari) in transmitting diseases remains relatively understudied (Reeves et al. 2006) and most research has been focused on rickettsialpox (caused by Rickettsia akari Huebner) and scrub typhus (caused by Orientia tsutsugamushi Tamura et al.) that are transmitted by the house mouse mite (Liponyssoides sanguineus Hirst, Dermanyssidae) and chigger mites (Trombiculidae), respectively (Elliott et al. 2019; Paddock 2020).

Mites of the genus Laelaps (Mesostigmata: Laelapidae) are common ectoparasites of small mammals, particularly rodents (Jameson 1965). Although some neotropical laelapines might not feed on blood (Martins-Hatano et al. 2002), other species such as Laelaps echidninus and Laelaps nuttalli are hematophagous parasites of humans and rodents (Wharton and Cross 1957; Jameson 1965; Krantz and Walter 2009) that might be capable of transmitting diseases among wild animals and humans (Wharton and Cross 1957; Valiente Moro et al. 2005). However, studies on Laelaps mites and their role as disease vectors remain very limited, despite some rickettsiae having recently been detected (though vector competence in pathogen transmission was not demonstrated) in Laelaps mites. These included Anaplasma platys and Anaplasma phagocytophilum in Laelaps spp. (Reeves et al. 2006), and Rickettsia helvetica and Rickettsia felis in Laelaps agilis (Miťková et al. 2015; Radzijevskaja et al. 2018).

Notably, studies of Laelaps mites in Taiwan remain scarce. The only research that has focused on this group of ectoparasites in Taiwan listed the rodent host and geographic occurrence of eight Laelaps species (Jameson 1965). Tsai et al. (2010) identified Bartonella species in shrews and rodents trapped near animal farms and their associated fleas and lice in Taiwan but did not detect Bartonella in 38 adult L. echidninus and 41 pools of nymphal Laelaps mites. Related studies in other Asian countries are also fragmentary. The host associations of Laelaps mites have been documented in Asiatic Russia (Vinarski and Korallo-Vinarskaya 2016), Bangladesh (Fuehrer et al. 2012), China (Guo et al. 2013), Indonesia (Durden and Page 1991), Laos and Vietnam (Allred 1970), and Malaysia (Ahamad et al. 2013).

Spotted fever group (SFG) rickettsioses are transmitted by ectoparasites infective of Rickettsia species. Hard ticks are the primary vectors, though some Rickettsia species are also transmitted by fleas and mites (Raoult and Roux 1997; Parola et al. 2005). Recent decades have witnessed the emergence of many novel Rickettsia species, whereas human cases of SFG rickettsioses have continuously appeared in previously unrecognized regions (Raoult and Roux 1997; Perlman et al. 2006; Parola et al. 2013).

Although Rickettsia species have previously been identified in Laelaps mites (Miťková et al. 2015; Radzijevskaja et al. 2018), similar investigations remain scarce, particularly in Asia. Tay et al. (2014) had attempted but failed to detect rickettsiae in Laelaps mites in Malaysia. The first aim of the present study was to detect Rickettsia-related species in Laelaps mites in Taiwan to assess their potential significance for transmitting rickettsiae. Moreover, knowledge regarding the relative contribution of distinct host species in supporting parasites is fundamental for disease mitigation by controlling certain hosts; therefore, the secondary aim of this study was to systematically investigate and compare the extent of Laelaps mite infestation among small mammal hosts in different parts of Taiwan. To our knowledge, this is the first study to detect Rickettsia in Laelaps mites in Taiwan while also systematically assessing their hosts. Importantly, the results of this study will contribute to the general understanding of Laelaps mites and their potential importance as disease vectors.

Materials and methods

Small mammal trapping and collection of mites

This present study was part of a research effort to investigate Rickettsia infections in small mammals and associated ectoparasites in rural lowland Taiwan (< 500 m in elevation, Kuo et al. 2015a, b; Wang et al. 2020; for more details on the trapping and handling of small mammals in this study, refer to Kuo et al. 2015a). Briefly, from 2006 to 2010, small mammals (shrews and rodents) were trapped at three counties in eastern Taiwan (Yilan: 121.7378° E, 24.7021° N; Hualien: 121.6112° E, 23.9911° N; Taitung: 121.1438° E, 22.7613° N), three counties in western Taiwan (Taoyuan: 121.2300° E, 24.9554° N; Taichung: 120.6736° E, 24.1477° N; Kaoping: 120.3014° E, 22.6273° N), and two counties located in Taiwan’s offshore islets (Kinmen: 118.4171° E, 24.4414° N, Penghu: 119.5833° E, 23.5833° N) (Fig. 1). In each county, 80 Sherman traps (26.5 × 10 × 8.5 cm) and 80 mesh traps (27 × 16 × 13 cm) were deployed in agricultural fields or rural villages for at least two different seasons (spring: March to May; summer: June to August; fall: September to November; winter: December to February) and baited with sweet potato smeared with peanut butter. Trapped small mammals were euthanized with an overdose of Zoletil 50 (Virbac SA, Carros, France). Laelaps mites were brushed off from hosts above a plastic pan until no mites could be found and collected with forceps, then preserved in 70% ethanol and stored at − 70 °C for subsequent molecular investigation. Small mammals and Laelaps mites were morphologically identified to species following Lin (1980) and Jameson (1965), respectively.

Fig. 1

Study sites for the trapping of shrews and rodents and collecting of associated Laelaps mites (Acari: Laelapidae) in lowland Taiwan from 2006 to 2010

Detection of Rickettsia species in mites

Laelaps mites were pooled for the detection of Rickettsia, with each pool containing five mites that typically belong to the same species and were collected from the same host individual. If this could not be achieved, mites of different species or from different host individuals were combined for a pool of five mites (approximately 20% of samples). For each county, mites of all present species were included in the assay except for one mite species with only two individuals, and each county included up to 10 pools. Mite DNA was extracted using a QIAamp DNA Mini Kit (Qiagen, Hilden, Germany) after sterilized mite specimens were ground in a solution containing 180 ml SPG buffer [3.0 mM KH2PO4 (Merck), 7.2 mM K2HPO4 (Merck), 40 mM l-glutamic acid (Sigma), 218 mM sucrose (Sigma)] and 20 ml proteinase K (Sigma), and vortexed at 56 °C overnight. Mites were detected of Rickettsia presence with nested polymerase chain reaction (PCR) following Kuo et al. (2015a), which targeted the genes encoding the 120- to 135-kDa outer membrane protein B (ompB; outer primer pair: ompB OF, 5′-GTA ACC GGA AGT AAT CGT TTC GTA A-3′; ompB OR, 5′-GCT TTA TAA CCA GCT AAA CCA CC-3′; inner primer pair: ompB SFG IF, 5′- GTT TAA TAC GTG CTG CTA ACC AA-3′; ompB SFG/TG IR, 5′- GGT TTG GCC CAT ATA CCA TAA G-3′; ompB TG IF, 5′-AAG ATC CTT CTG ATG TTG CAA CA-3′; 426 bp) and citrate synthase (gltA; outer primer pair: RpCS.877p, 5′-GGG GGC CTG CTC ACG GCG G-3′; RpCS.1258n, 5′-AAT GCA AAA AGT ACA GTG AAC A-3′; inner primer pair: RpCS.896, 5′-GGC TAA TGA AGC AGT GAT AA-3′; RpCS.1233n, 5′-GCG ACG GTA TAC CCA TAG C-3′; 338 bp). Laboratory Rickettsia rickettsii DNA and DEPC-treated water were used as positive and negative controls, respectively. R. rickettsii DNA was prepared from R. rickettsii antigens coated in a R. rickettsii IFA substrate slide (product code IF0101; Focus Technologies) and extracted using a QIAamp DNA Mini Kit. Samples were deemed positive for Rickettsia presence when either the ompB or gltA gene was detected. Rickettsia species in positive samples were identified by purifying the PCR products using a QIAquick Gel Extraction Kit and then sequencing once in each direction. The sequencing was performed by a commercial company (Mission Biotech, Taipei, Taiwan). DNA nucleotide sequences were assessed using the Basic Local Alignment Search Tool (http://www.ncbi.nlm.nih.gov) for any resemblance to known Rickettsia species. However, due to the limited number of genes detected and the very small fragments sequenced, only broad category of rickettsiae (e.g. SFG rickettsiae) instead of precise Rickettsia species was reported in this study.

Data analysis

The difference in mite load (mean mites per host individual) among host species, study sites, and seasons was investigated with a negative binomial generalized linear model (NBGLM) to account for overdispersion of data, and a significant difference was evaluated based on the 95% Wald confidence interval. These procedures were implemented in SPSS Statistics v.19.0 (IBM Corporation). We used EpiTools Epidemiological Calculators (Sergeant 2018) to estimate mean and 95% confidence interval (CI) of individual-level (per mite) prevalence of Rickettsia infection in mites with a frequentist approach assuming perfect test, with confidence intervals calculated based on binomial theory following Cowling et al. (1999).

Results

Host association of Laelaps mites

We captured a total of 1,004 small mammals that included one shrew and ten rodent species (Table 1). The most common species was Rattus losea (44.9% of total mammals captured), followed by Suncus murinus (21.1%) and Mus caroli (13.3%). Rattus losea also hosted the highest percentage of mites (76.6% of total mites), followed by M. caroli (11.5%), whereas only 0.8% of mites were collected from S. murinus despite the commonness of this shrew species (Table 1). Upon excluding the non-representative Niviventer coninga due to its small sample size (two individuals), the exotic Rattus exulans exhibited a higher mite load (an average of 1.76 mites per individual) than most other species except for Rattus norvegicus (mean = 1.70), R. losea (1.45), and M. caroli (0.73) (NBGLM, all P < 0.05). Among all host species, R. exulans also had a higher prevalence of infestation (52.0%) (Table 1).

Table 1

Total number of captures of each shrew and rodent species at eight study sites in lowland Taiwan from 2006 to 2010, as well as prevalence of Laelaps mite infestation (% infested individuals), Laelaps mite load (mean [± SE] no. mites per host individual), total Laelaps mites collected, and Laelaps mite species identified from each host species

Mammal species	No. mammals (% of total)	Prevalence of mite infestation (%)	Mite load*	No. mites (% of total)	Infested mite species (no. mites)
Shrew
Suncus murinus	212 (21.1)	2.4	0.03 ± 0.02a	7 (0.8)	L. echidninus (2); L. nuttalli (5)
Rodent
Apodemus agrarius	20 (2.0)	15.0	0.35 ± 0.25ad	7 (0.8)	L. echidninus (2); L. nuttalli (5)
Bandicota indica	55 (5.5)	7.3	0.09 ± 0.05a	5 (0.6)	L. nuttalli (5)
Callosciurus erythraeus	2 (0.2)	0	0b	0 (0)
Mus caroli	134 (13.3)	25.4	0.73 ± 0.17cd	98 (11.5)	L. nuttalli (93); L. sedlaceki (1); Laelaps spp. (4)
Mus musculus	45 (4.5)	6.7	0.07 ± 0.04a	3 (0.4)	L. nuttalli (3)
Niviventer coninga	2 (0.2)	50.0	5.00 ± 5.00ac	10 (1.2)	L. echidninus (6); L. sedlaceki (4)
Rattus exulans	25 (2.5)	52.0	1.76 ± 0.60c	44 (5.2)	L. echidninus (10); L. nuttalli (31); L. sedlaceki (3)
Rattus losea	451 (44.9)	29.3	1.45 ± 0.19c	653 (76.6)	L. echidninus (181); L. nuttalli (288); L. sedlaceki (179); L. myonyssognathus (2); Laelaps spp. (3)
Rattus norvegicus	10 (1.0)	30.0	1.70 ± 1.48cd	17 (2.0)	L. echidninus (4); L. nuttalli (4); L. sedlaceki (9)
Rattus tanezumi	48 (4.8)	10.4	0.19 ± 0.10a	9 (1.1)	L. echidninus (1); L. nuttalli (7); L. sedlaceki (1)
Total	1004	20.2	0.85 ± 0.09	853	L. echidninus (206); L. nuttalli (441); L. sedlaceki (197); L. myonyssognathus (2); Laelaps spp. (7)

*Means followed by a different letter are significantly different (negative binomial generalized linear model: P < 0.05)

Laelaps mite fauna and geographical and seasonal variation

A total of 853 Laelaps mites of at least four species were collected (Table 2). Laelaps nuttalli was the most abundant species, constituting over half of the collected mites (51.7%). This species was followed by equally abundant L. echidninus and Laelaps sedlaceki, each accounting for approximately one-quarter of the mites (24.2 and 23.1%, respectively). In contrast, Laelaps myonyssognathus was rare, with only two individuals collected (0.2%). The three most common species (L. echidninus, L. nuttalli, and L. sedlaceki) can be found on most small mammal species (seven, nine, and six, respectively), whereas L. myonyssognathus was only collected from R. losea (Table 1).

Table 2

Total number of Laelaps mites and mite load (mean [± SE] no. mites per host individual) on the primary host Rattus losea collected from shrews and rodents at each study site of Taiwan from 2006 to 2010

Study site	Mite species						Mite load on R. losea (no. hosts)*
	L. echidninus	L. nuttalli	L. sedlaceki	L. myonyssognathus	Laelaps spp.	Overall
Eastern Taiwan
Yilan	18	14	18	0	0	50	0.43 ± 0.14 (91)a
Hualien	19	87	13	0	0	119	2.05 ± 0.89 (19)b
Taitung	25	46	19	0	0	90	3.05 ± 1.28 (19)b
Western Taiwan
Taoyuan	43	66	97	1	4	211	1.57 ± 0.42 (116)b
Taichung	19	22	32	0	0	73	1.77 ± 0.53 (31)b
Kaoping	8	30	12	1	3	54	0.59 ± 0.21 (44)a
Offshore islets
Kinmen	28	77	6	0	0	111	1.48 ± 0.38 (75)b
Penghu	46	99	0	0	0	145	2.55 ± 0.88 (56)b
Overall	206	441	197	2	7	853	1.45 ± 0.19 (451)

*Means followed by a different letter are significantly different (negative binomial generalized linear model: P < 0.05)

Geographically, L. nuttalli and L. echidninus were generally common in different parts of Taiwan. On the other hand, L. sedlaceki occurred mainly on the main island of Taiwan and was uncommon on offshore islets. Moreover, the rare L. myonyssognathus was observed in western Taiwan only (Table 2). The mite load on R. losea, the primary small mammal host, varied considerably among regions (NBGLM, Wald’s χ2 = 86.4, P < 0.001; range 0.43 to 3.05), with lower loads occurring in Yilan (mean = 0.43) in eastern Taiwan and Kaoping (0.59) in western Taiwan (all P < 0.05) (Table 2). There was also a seasonal variation in mite load on R. losea (NBGLM, Wald’s χ2 = 17.5, P < 0.005), with higher and lower loads in the fall (mean = 2.04) and winter (0.62), respectively (Fig. 2a). Likewise, load of L. echidninus and L. nuttalli was generally higher and lower in the fall and winter, respectively (Fig. 2b, c); however, load of L. sedlaceki was higher in the spring but lower in the fall (Fig. 2d).

Fig. 2

Seasonal difference in the mean (+ SE) number of Laelaps mites collected from Rattus losea in lowland Taiwan from 2006 to 2010. a Mites of all species; b Laelaps echidninus; c Laelaps nuttalli; d Laelaps sedlaceki. Means within a panel capped with different letters are significantly different, based on the 95% Wald confidence interval

Detection of in mites

In total 72 pools of Laelaps mites were examined for the presence of Rickettsia. The positivity rates per pool were 4.2 and 9.7% based on the ompB and gltA genes, respectively (Table 3), whereas positivity rates per mite were 0.85% (95% CI 0.17–2.46%) and 2.02% (0.81–4.13%), respectively. The positivity rate was higher in Taoyuan and Penghu for the ompB gene, and it was higher in Hualien, Taitung, and Kinmen for the gltA gene. For the ompB gene, three DNA sequences that matched the closest in GenBank (99.0–99.2% similarity) to SFG rickettsiae were each detected from L. echidninus, L. sedlaceki, and a pool of L. echidninus and L. nuttalli (GenBank Accession Numbers MT611477–MT611479); for the gltA gene, four sequences with the highest similarity to SFG rickettsiae (99.3–100% similarity) were each detected from L. echidninus, L. nuttalli (2×), and L. sedlaceki (MT611473-MT611476) (Table 3). All of the mites with Rickettsia species detected were collected from R. losea.

Table 3

Positivity rate for Rickettsia species infection and Rickettsia species identification in pools of Laelaps mites for each study site in Taiwan from 2006 to 2010

Study site	Positivity rate (%) for ompB gene	Rickettsia spp. detected (no. detections from mite species)	Positivity rate (%) for gltA gene	Rickettsia spp. detected (no. detections from mite species)	Laelapine mite species assayed (no. pools)
Eastern Taiwan
Yilan	0 (0 out of 8)	–	0 (0 out of 8)	–	L. echidninus (2); L. nuttalli (1); L. sedlaceki (1); L. echidninus + L. sedlaceki (4)
Hualien	0 (0 out of 10)	–	10 (1 out of 10)a	–	L. nuttalli (6); L. echidninus + L. nuttalli (1); L. echidninus + L. sedlaceki (2); L. echidninus + L. nuttalli + L. sedlaceki (1)
Taitung	0 (0 out of 10)	–	30 (3 out of 10)	SFG rickettsiaeb (1 L. echidninus; 1 L. nuttalli; 1 L. sedlaceki)	L. echidninus (3); L. nuttalli (4); L. sedlaceki (1); L. echidninus + L. sedlaceki (1); L. nuttalli + L. sedlaceki (1)
Western Taiwan
Taoyuan	20 (2 out of 10)	SFG rickettsiae (1 L. echidninus; 1 L. sedlaceki)	0 (0 out of 10)	–	L. echidninus (2); L. nuttalli (3); L. sedlaceki (5)
Taichung	0 (0 out of 7)	–	0 (0 out of 7)	–	L. echidninus (2); L. nuttalli (2); L. sedlaceki (3)
Kaoping	0 (0 out of 7)	–	0 (0 out of 7)	–	L. nuttalli (3); L. sedlaceki (1); L. echidninus + L. nuttalli (2); L. echidninus + L. sedlaceki (1)
Offshore islets
Kinmen	0 (0 out of 10)	–	30 (3 out of 10)a	SFG rickettsiae (1 L. nuttalli)	L. echidninus (3); L. nuttalli (6); L. sedlaceki (1)
Penghu	10 (1 out of 10)	SFG rickettsiae (1 L. echidninus + L. nuttalli)	0 (0 out of 10)	–	L. echidninus (2); L. nuttalli (7); L. echidninus + L. nuttalli (1)
Overall	4.2 (3 out of 72)	SFG rickettsiae (3)	9.7 (7 out of 72)	SFG rickettsiae (4)

Each pool contained five mites typically belonging to the same species and collected from the same host individual, or mites of different species or from different host individuals were combined for a pool of five mites. These mites with Rickettisa species detected were all collected from Rattus losea

aSome Rickettsia detected but could not be sequenced successfully

bSpotted fever group rickettsiae

Discussion

In addition to the R. felis and R. helvetica that were detected in L. agilis in Europe (Miťková et al. 2015; Radzijevskaja et al. 2018), we identified Rickettsia species closely related to the SFG rickettsiae in Laelaps mites for the first time in Asia. Moreover, these Rickettsia species were largely detected in L. echidninus and L. nuttalli, which are known to bite humans (Wharton and Cross 1957). However, due to that only fragments of gltA and ompB genes were identified, further definitive confirmation of Rickettsia species is required. Additionally, SFG rickettsiae are primarily transmitted by ticks (Parola et al. 2013). Therefore, it is more likely that the rickettsiae detected in the present study were contained in blood meals ingested by the mites rather than mites being infective and capable of transmitting these pathogens. For example, prevalence of Rickettsia infection in the small mammal hosts during our parallel studies on Rickettsia detection in Taiwan was high (Kuo et al. 2015a) although these Rickettsia-positive mites were not derived from the same small mammal hosts examined for the presence of Rickettsia. The non-concordance in detection of SFG rickettsiae based on gltA and ompB genes, respectively, also suggests that the concentration of rickettsiae in these mites might be too low for successful amplification by both gene targets simultaneously. In this context, it cannot be excluded that Laelaps mites might define alternative transmission routes for some of these Rickettsia species. For example, although fleas are the primary competent vectors of R. felis, this pathogen can also be successfully cultivated in tick and mosquito-derived cell lines (Reif and Macaluso 2009). Therefore, the presence of tick-borne rickettsiae warrants further investigation on the competence of hematophagous Laelaps mites in transmitting Rickettsia species, which has never been investigated to date.

A total of eight Laelaps mite species have previously been recorded in Taiwan (Jameson 1965). These include the four species that were also collected for the present study in lowland areas (< 500 m in elevation; Laelaps prognathous in Jameson (1965) considered a junior synonym of L. myonyssognathus by Mitchell (1968)). The other four species (L. agilis, Laelaps clethrionomydis, Laelaps traubi, and Laelaps turkestanicus) were collected from rodents residing at elevations > 900 m (Jameson 1965); therefore, it is not surprising that they were not discovered in the present study.

Among the four species, L. nuttalli was the most abundant, observed in all study sites, and collected from most small mammal species—except from the uncommon Callosciurus erythraeus and N. coninga (Tables 1 and 2). In Taiwan, L. nuttalli can reach an elevation of 900 m (Jameson 1965). L. nuttalli is also a common ectoparasite of commensal rodents in most parts of the world, including Africa, America, Asia, and the Indian Ocean and Pacific islands, being distributed from sea level to > 3000 m in elevation (Allred 1969). This species has also been found on bats (Ahamad et al. 2013). In the present study, Rickettsia species with the highest similarity to SFG rickettsiae have been detected in L. nuttalli which is also closely associated with commensal rodents that have frequent contact with humans, thus highlighting that the medical importance of this widely distributed mite species is worthy of further investigation. Similarly, L. echidninus is also a cosmopolitan ectoparasite of commensal rodents (Durden and Page 1991). Jameson (1965) only found this species on introduced or domestic Rattus rats but not on native rodents. Instead, we identified L. echidninus on seven host species, including N. coninga, which is native and endemic to Taiwan, thereby revealing a wider host range than previously considered. Likewise, L. sedlaceki was previously found only on N. coninga (Jameson 1965); however, it was also collected from another five host species in the current study (largely Rattus rats), which is similar to what has been observed in Borneo and New Guinea (Strandtmann and Mitchell 1963; Ng et al. 2017). In addition, species with the highest resemblance to SFG rickettsiae have also been identified in both L. echidninus and L. sedlaceki, once again underscoring the importance of assaying their competence for disease transmission. Lastly, L. myonyssognathus was the least common species in the present study, being found only on the abundant R. losea in western Taiwan, which is consistent with previous research (Jameson 1965); however, its hosts included at least nine shrew and rodent species in its distribution from East to South Asia (Mitchell 1968).

Over 75% of Laelaps mites were collected from R. losea, thus indicating that this common rodent is important in sustaining Laelaps mite populations. We have also identified R. losea as the primary host of ticks, chigger mites, and fleas in lowland Taiwan (Kuo et al. 2015b, c; Wang et al. 2020). Together, these studies demonstrate the prominent role of R. losea in supporting arthropod disease vectors in rural lowland Taiwan, which suggests that this species could be targeted when disease control measures are required. Interestingly, we also observed that R. exulans had a higher Laelaps mite burden similar to what has been reported in Indonesia (Durden and Page 1991) and Hawaii (Mitchell 1964). In Taiwan, R. exulans is an invasive species that is currently present only in Hualien (eastern Taiwan). This locally abundant rodent was found to serve only a minor role in supporting chigger mites due to its low chigger mite burden (Kuo et al. 2011). On the contrary, the higher Laelaps mite burden identified in this study suggests that the invasive R. exulans could be a competent host and its presence could augment the population of Laelaps mites, which subsequently increases the mite burden of other coexisting native rodents; such so-called “parasite spillback” effect can arise when invasion of a competent host increases the parasite load in native hosts by supporting a large number of local parasites (Kelly et al. 2009). Therefore, further validation of the importance of R. exulans in this context is recommended.