Molecular detection of Orientia tsutsugamushi in ectoparasites & their small mammal hosts captured from scrub typhus endemic areas in Madurai district, India

Geographically confined areas of Asia-Pacific region consisting of south and south-east Asia, northern Australia, Pacific and Indian ocean islands are endemic for scrub typhus infection1. There were more than one million scrub typhus cases annually in Asia with high mortality rate as reported in 20152. Central nervous system infection has high mortality rate34. The emergence of scrub typhus has been reported from various parts of the world, including Chile and Africa5. Re-emergence of scrub typhus in various States of India with diverse ecological parameters has been reported6. Scrub typhus infection is presented as non-specific flu-like symptoms, including fever, cough, rash, myalgia, vomiting, nausea, eschar at the site of bite, abdominal pain and generalized lymphadenopathy. About one-third of infections proceed to multi-organ failure, including renal, myocardial, septic shock, meningoencephalitis and pulmonary disturbances6. It is also reported that diagnosis is complicated as eschar may not manifest in all cases4. Case fatality rate (CFR) of scrub typhus differs across various countries and regions. The CFR can increase to 30-70 per cent if treatment is not initiated promptly12.

Orientia tsutsugamushi (formerly Rickettsia tsutsugamushi78), the causative agent of scrub typhus infection, is transmitted to mammalian hosts (including humans) by the larvae stage of Leptotrombidium mites, also called chiggers9 which serve as its primary reservoirs. The infected larval mite feeds on vertebrates, including humans, rodents, and other small mammals. The nymph and adult mites reside in the soil. Rodents infected with O. tsutsugamushi can maintain the infection for a greater time period5.

The diagnosis of scrub typhus in humans is based on the patient’s history (e.g., travel, work including farming) and clinical features (like the presence of eschar) and confirmation by the standard laboratory methods, including immunofluorescence assay, enzyme-linked immunosorbent assay and Polymerase Chain Reaction (PCR) (direct, nested and real-time)10. The nested PCR-based method is more reliable and easier for diagnosis. As other Rickettsia sp., can also cause spotted fever, typhus fever which can mimic scrub typhus-like infection; it is essential to differentiate the infection at the earliest to avoid misdiagnosis and treatment11. GroEL chaperonin, (60 kDa heat-shock protein), is typically used for diagnostics, as this gene is highly conserved in O. tsutsugamushi strains. GroEL gene sequences are known to have higher degree of divergence across the rickettsiae; and have been shown to facilitate the rapid diagnosis of rickettsial diseases as well as differentiate between these as against other acute febrile diseases1213.

So far only a few reports are available on the prevalence of O. tsutsugamushi in chigger mites on domestic rodents/shrew1415161718 thus it is important to monitor small animals, including rodents and shrews, in the transmission of scrub typhus. Recently, we have reported the presence of trombiculid mites, which are predominant vectors for O. tsutsugamushi in domestic rodents in Madurai1920. In continuation of the same, in the present study we undertook to detect O. tsutsugamushi by nested PCR based on the groEL gene expression in mites from Acari and fleas from ectoparasites collected on domestic rodents, shrews, dogs and cats in Tamil Nadu. This study (i) helps in understanding the presence of scrub typhus pathogens in both vector and vertebrate host, (ii) it also helps to identify potential high-risk areas for scrub typhus infection.

Material & Methods

Study sites: Nine study sites were selected within Madurai district and grouped into urban (Usilampattii, Thirumangalam and B.B. Kulam), sub-urban (Keelaiyur, Sholavandan and Peraiyur), and rural (Chatrapatti, Katchiakatti and Vadapalanji) habitats. The Madurai district is situated in southern part of Tamil Nadu State, lies between 9°33’30” N to 10°18’50” N Latitude, 77°29’10” E to 78°28’45” E Longitude with an area spanning across 3710 km² (https://madurai.nic.in/district-profile/). Previously Varghese et al21 reported scrub typhus infection in the areas near Madurai district. In order to understand the prevalence of O. tsutsugamushi in Madurai district, in this study entomological and small mammal surveillance was undertaken. The study was carried out between July 2017 to June 2018 after seeking approval by the Institutional Animal Ethics Committee of Madurai Medical College, Madurai, Tamil Nadu.

Trapping of rodents and shrews: Small mammals were trapped over a period of one year using 360 Sherman traps for each habitat (urban, semi urban and rural ). A total of 1080 traps were placed in all habitats i.e., 324 traps were placed indoors (urban-64, sub urban -103, and rural-157) and 756 traps were placed outdoors (urban-296, sub urban-257, and rural-203) including indoors (urban -64, sub-urban -103 and rural -157) and outdoors. Traps were kept at indoor and outdoor households in residential areas before dusk (5-6 PM) hours and withdrawn after dawn time (6-7 AM), i.e. next morning. The rodents/shrews were attracted by eatables (potato chips/wheel chips) fried in coconut oil and placed within the Sherman traps22. The trapped rodents/shrews were brought safely to the laboratory by keeping them inside separate cloth bags.

Processing of small mammals: The trapped rodents/shrews were euthanized with thiopentone sodium injection (60 mg/kg intravenously). The euthanized rodent and shrews were then taken out of the trap1520. A cardiac puncture was performed to collect 1-2 ml blood using the appropriate procedure in ethylenediamine tetraacetic acid tubes and stored at −20°C until further use15. Euthanized rodents/shrews were identified up to species level using external morphological characteristic features2324.

Collection of chigger mites from rodents/shrews: Trombiculid mites were collected using a fine brush from the rodents and shrews. Ear pinna and femur were the major sites for the collection of mites. The collected mites were stored in 70 per cent ethanol2022.

Tick collection from rodents/shrews: Ticks were collected from the rodents/shrews using curved tweezers from the body areas such as the head, ear, face and legs, which are major sites for tick inhabitation. The ticks were grabbed using a curved tweezer as close as possible to the skin surface, then gently pulling away from the skin without twisting at a 45° angle. The collected ticks were transferred and stored in vials containing 70 per cent ethanol.

Collection of fleas from cat and dog: A total of 100 households were selected at random from rural, sub-urban and urban habitats, for testing the presence of fleas on cats and dogs. About 20 dogs and cats were tested at every habitat. Fleas were manually collected in a white sheet/white enamel tray by combing the body hair of the host. The fleas were then immediately transferred into vials containing 70 per cent ethanol with a pair of forceps.

Identification, dissection and preservation of chiggers: Each chigger mite was placed in a drop of 70 per cent ethanol solution on a slide. Under a dissection microscope, the exoskeleton of the mite was punctured, the idiosoma and internal tissue contents including haemolymph were squeezed out using angled or notched forceps. The internal contents of the chigger in ethanol were thoroughly broken into fragments using a minute pin. A tiny amount of the internal content suspension was transferred to an Eppendorf tube using a micropipette. The tube was kept in a deep freezer (−20°C) until tested. A new sterile dissecting needle was used for each chigger mites.

The exoskeleton of the chigger, which remained mostly intact, was transferred to lactophenol solution (50 ml lactic acid, 25 ml phenol and 25 ml distilled water) for clearing the specimen. All mite specimens were then mounted on separate slides with Hoyer’s medium and identified upto species level under Nikon ECLIPSE (E200) microscope2526272829303132. These mounted slides and other preserved specimen vials were deposited in National Mite Museum, under the Unit of Vector-Borne and Zoonotic Diseases, ICMR-Vector Control Research Centre Field Station Madurai, Tamil Nadu.

Chigger index and chigger infestation rate: Chigger index (CI) was calculated by dividing the total number of chigger mites collected by the total number of hosts examined. Similarly, the chigger infestation rate (CIR) was calculated as the percentage of the total number of chigger mites collected divided by a total number of hosts collected with chigger mites22.

Molecular detection of Orientia tsutsugamushi in ectoparasites, rodents and shrew DNA extraction of ectoparasites and small mammals: The internal suspension content (see above) collected from species-wise pooled chigger (10-15 Nos.), adult mites, ticks, fleas and individual sera samples of rodents and shrews were used for DNA extraction (Table I). DNA was extracted using a QIAmp blood and tissue mini kit (Qiagen, Hilden, Germany) as per the manufacturer’s instructions. The DNA was then stored at 4°C and used as templates for PCR and stored at −20°C for later.

Nested PCR amplification for Rickettsia and Orientia tsutsugamushi identification: The initial primer used for the first PCR was based on the groEL gene sequence and target agents of the spotted fever, typhus group of rickettsiae and O. tsutsugamushi. The primers were Gro1: (5’-AAGAAGGACGTGATAAC-3’; position from 603-618 bp); Gro2: (5’-ACTTCACGTAGCACC-3’; position from 1251-1238)33. Thirty microliter of reaction mixture contained 15 µl Taq 2X Master Mix red (Ampliqon), 0.3 µl of each primer (Gro1 and Gro2), 5 µl of DNA template and 9.4 µl of distilled water. PCR reactions were performed in Thermal cycler ABI PCR (Applied Biosystems, USA). The nested PCR used two pairs of inner primers33 in a single reaction. The first pairs of nested primers (SF1 and SR2) target spotted fever and typhus group of rickettsia (217 bp segment), while second pairs of inner primers (TF1 and TR2) target O. tsutsugamushi (364 bp segment) (Supplementary Table I). The 30 µl nested PCR reaction mixture contains 10 µl Taq 2x Master Mix red (Ampliqon), 0.3 µl of each primer (TF1, TR2, SF1 and SR2), 3 µl of first-round products as a template and 15.8 µl of distilled water. Nested PCR reactions were followed by an initial denaturation step at 95°C for two minutes and 40 cycles of denaturation at 95°C for 35 seconds, annealing at 55°C for 35 seconds and 72°C for 35 seconds and final extension at 72°C for five minutes.

Electrophoresis and sequencing: PCR products subjected to electrophoresis at 90 V for 60 min on two per cent agarose gel with 0.5 µg/ml ethidium bromide and visualized on gel documentation system (Bio-Rad). Further, the PCR products were gel purified (Invitrogen, USA) and Sanger sequenced bidirectionally (Genurem Biosciences, Tamil Nadu). The obtained sequences were assembled (DNASTAR) and submitted to NCBI.

Phylogenetic analysis: Gene alignment was done along with reference sequences from GenBank by Clustal Omega and the phylogenetic analysis was performed using the maximum likelihood (ML) model in Randomized Axelerated ML (RAxML)34. A general time-reversible model of nucleotide substitution35 with a gamma correction for among-site rate variation and an estimated proportion of invariant sites was used. Internal nodes were statistically supported through boot strapping with 1000 replicates. The genetic distance (p) within and between groups was calculated using MEGA 7.036.

Statistical analysis: The Chi-square test was performed using SPSS version 25 (IBM Corp; NY, USA).

Results

During the 12 months (July 2017-June 2018) study period, 330 animals from three habitats comprising Rattus rattus (95), R. norvegicus (3), Mus musculus (12), Suncus murinus (32), Bandicota bengalensis (6), Tatera indica (3), Canis familiaris (domestic dog, 153) and Felis catus (cat, 26) were screened for ectoparasites infestation. All animals with the exception of T. indica were trapped in all habitats; interestingly, T. indica was trapped only in rural sites. Among the three habitats, individual genus wise, there was no significant difference between Rattus rattus, (χ²=0.5171, df-2, P>0.05) and Suncus murinus, (χ²=0.1525, df-2, P >0.05 with the exception of C. familaris (χ²=5.9983, df-2, P <0.05, n=330) (Table I).

A total of 3860 ectoparasites (Table II) were collected from 330 animals, where 2741 (71%) ectoparasites were morphologically identified (Table III). Interestingly, majority of the ectoparasites, including chigger mites (3151/3219-97.8%), adult mites (6/6-100%), ticks (10/10-100%) and fleas (515/626-82.2%) were obtained from outdoor rural habitat (Table II). Among the identified 2741 ectoparasites from animal hosts, 2099 were chigger mites (belonging to L. deliense, L. indicum, L. keukenschrijveri, L. rajesthanensis, Schoengatiella ligula, Schoengastia sp., Neotrombicula microti, Microtrombicula sp., and Trombicula hypodermata), six were adult mites (belonging to Oribatida sp. and D. gallinae), 626 were fleas (belonging to X. astia, X. cheopis, C. felis and Ctenophalides sp.) and 10 were ticks (belonging to Rhipicephalus sanguineus and Rhipicephalus haemaphysaloides) (Tables II and III). Among 151 trapped rodents/shrews, 57 (37.75%) rodents/shrews were positive for trombiculid mites. M. musculus did not harbour trombiculid mite (Table III).

There was no month-wise significant difference among the different habitats studied for the CI (F-0.1431, df-35, P>0.05) and CIR (F-0.0561, df-35, P >0.05) (Table IV). Chigger density was influenced by the climate in each month; according to temperature, rainfall and relative humidity, the months were grouped into four seasons, i.e. southwest monsoon (June-September), northeast monsoon (October-November), winter (December-February) and summer (March-May). The winter season was dominated in all three habitats with high CI and CIR. The overall chigger collection from all the habitats was higher during cooler months, including winter and northeast monsoon, was 2377 (74%) in comparison to other months, which was about 835 (26%), showing a significant difference between cooler months and other months in the chigger collection (t-3.660, df-10, P <0.05) (Table IV).

Orientia tsutsugamushi prevalence in small mammals and ectoparasites: The prevalence of O. tsutsugamushi was identified by amplification of the groEL gene. Due to logistical difficulties in examining O. tsutsugamushi in all chigger mites and other ectoparasites, a representative sample of 10-50 per cent was used, i.e. urban [416 (19.7%)], sub-urban [648 (30.92%)] and rural [1035 (49.38%)] habitat. Of 2099 chiggers from all three habitats, 2081 (99.14%) chiggers were collected outdoors, which shows its abundance in outdoors alone (Table II). None of the chigger mites were positive for O. tsutsugamushi (Table V). Among the three habitats, there was no significant difference in O. tsutsugamushi positivity observed among the ectoparasites collected from indoors (χ²=0.9472, df -2, P >0.05) and outdoors (χ²=3.9679, df-2, P >0.05) host. There was also no significant difference (χ²=2.777, df-2, P >0.05) observed in O. tsutsugamushi-positive ectoparasite among three different habitats. Among the adult mites, the presence of O. tsutsugamushi was observed in Oribatida (1/1-100%) and in D. gallinae (1/2-50%). Among 5 larval ticks, 35 fleas and sera of 6 species of rodents/shrew tested; 3 ticks, 4 fleas and sera of 3 rodents/shrew were positive for O. tsutsugamushi (Table V). Among the 47 fleas collected from F. catus, two are positive for O. tsutsugamushi (Tables V and VI). There was no correlation between chiggers (pooled) and O. tsutsugamushi, but there was a correlation between ectoparasites and positive host sera samples (Tables V and VI). None of the samples were positive for spotted fever and typhus group.

The nucleotide sequences obtained in this study were deposited in NCBI and assigned with the accession number HG995432 to HG995443. A total of 12 nucleotide sequences were submitted (Table VI). The sequences were confirmed to be from O. tsutsugamushi by BLAST with the groEL of the known reference sequence (NC_010793). The coding sequences of the obtained sequence were compared with groEL protein reference sequence WP_012460965 and found to be closely related to the conserved region (Fig. 1). Interestingly, the sequence of O. tsutsugamushi from R. rattus blood (HG995440) and its ectoparasite R. haemaphysaloides (HG995443) was similar (

Close

Fig. 1

The figure shows the groEL sequence similarity between our strains to the reference sequence. (A) The nucleotide sequences (HG995432-HG995443) showing 95-100 per cent similarity with 32-79 per cent coverage with the Ikeda sequence (NC_010793/AP008981). The red vertical lines indicate the difference in sequence within the grey horizontal lines. The green and red horizontal line indicates the gene and corresponding protein, respectively. (B) The protein sequence of our strains shows high similarity (71-100%) with reference sequence WP_012460965. The amino acids are given within the grey lines. The conserved regions are shown in red. The black horizontal lines show functional gene and protein. The figure is generated using the Multiple Sequence Alignment Viewer.

Export to PPT

Phylogenetic analysis: There were 174 groEL nucleotide sequences available in the NCBI database; shotgun and repeated sequences were omitted for phylogenetic analysis. Thus, the phylogenetic analysis was performed with the groEL gene from 166 sequences of O. tsutsugamushi and, Rickettsia japonica was used as an outgroup (Supplementary Table II). Based on the phylogenetic analysis, the groEL sequence clustered into four distinct clades supported by 100 per cent bootstrap. All O. tsutsugamushi identified in the study are grouped within clade IV (

Discussion

Rickettsial diseases, especially scrub typhus, is an emerging infectious disease and reported in several parts of India15. Rodents and shrews make an active role in the infestation of chiggers, flea and ticks to transmit rickettsial diseases203738. Scrub typhus cases were reported in the human population around Madurai21. In our study, we have observed O. tsutsugamushi in ticks (R. haemaphysaloides, R. sanguineus), fleas (X. astia, X. cheopis) and in non-trombiculid mites (D. gallinae and Oribatida sp.) from rodents and C. felis from F. cattus. As per our knowledge, this is the first-ever report of its kind in these species.

Ticks and fleas: Ticks are reported as carriers of O. tsutsugamushi1737383940. Ticks, especially Haemaphysalis hystricis, H. flava41 Ixodes spp.17 and Ixodes granulatus4243 are said to be vectors for O. tsutsugamushi. In our study, we have found O. tsutsugamushi in two tick species, R. sanguineus and R. haemaphysaloides. Fleas are ineffective as vectors of O. tsutsugamushi; however, it can maintain live pathogens for 11 days and transmit by biting44. However till date, there are no reports on fleas as a vector for the transmission of scrub typhus in humans. Chareonviriyaphap et al16 tested nearly 504 specimens of fleas (X. cheopis) for O. tsutsugamushi and none were positive. Similarly, O. tsutsugamushi was reported to be absent in C. orientis, C. f. felis and C. canis45. Howerver, in this study, O. tsutsugamushi was observed in X. cheopis, X. astia, Ctenophalides sp. and C. felis. However, it is not evident from this study whether these species will act as vectors for scrub typhus, the future study will be carried out in this regard.

Trombiculid ectoparasites: Trombiculid ectoparasites are the major vectors in transmitting O. tsutsugamushi. Majority of the reports show that O. tsutsugamushi was tested positive in trombiculid mites, particularly in the genus Leptotrombidium followed by Eutrombicula wichmanni, Odontacarus sp., M. chamlongi, Neotrombicula japonica and Helenicula miyagawai46. Leptotrombidium deliense was the most abundant species of trombiculid mite and it was reported as the predominant vector for scrub typhus infection in India and other countries646. Two hundred and four species of trombiculid mites were reported from India and four chigger mites, namely S. ligula, L. deliense, L. subintermedium and L. dihumerale were considered as the vectors for scrub typhus transmission26. In our study, we could collect only trombiculid mites L. deliense and S. ligula from the sites and unfortunately, they were negative for scrub typhus pathogens.

Non-trombiculid ectoparasites: Recently, non-trombiculid ectoparasites have been reported as vectors for O. tsutsugamushi46. Non-trombiculid ectoparasites such as Echinolaelaps echidninus43 Laelaps turkestanicus43 and Ornithonyssus bacoti141517 are said to host O. tsutsugamushi. Interestingly in this study, O. tsutsugamushi was found for the first time in non-trombiculid ectoparasites Oribatida sp., and D. gallinae.

Rodents/shrew: Several reports are available for O. tsutsugamushi infection in small mammals1647484950. R. rattus1516474849 S. murinus15 R. norvegicus1650 B. indica164849 R. bukit, R. argentiventer, R. berdmorei, R. losea, R. koratensis, B. savilei, R. exulans48 and Tupaia glis4748. In most cases, the prevalence rates were very low, i.e. ≤1 per cent15164950 with the exception where of Coleman et al48 and Frances et al47 reported 1-25 per cent prevalence rate. In this study, we detected O. tsutsugamushi in the blood of R. rattus (1/95-1.05%), S. murinus (1/32-3.13%) and B. bengalensis (1/6-16.7%). Interestingly, O. tsutsugamushi from R. rattus (

Phylogenetics of Orientia tsutsugamushi: GroEL gene has been used to distinguish scrub typhus from other rickettsial diseases13. Till date, O. tsutsugamushi genotypes have been reported from Japan (Kato, Hirano, Kuroki, Shimokoshi, Ikeda, Yamamoto, Kawasaki, Saitama), Thailand (TA678, TA763, TA716, TH1817), New Guinea (Karp, Kostival, Buie), South Korea (Boryong, Yonchon), Assam-Burma Border (Gilliam), Russia (B15), Australia (Litchfield) and Philippines (Volner)51 and India (IHS2252). According to Enatsu et al53 there were 31 serotypes (later called genotypes) based on the presence or absence of a 56 kDa type-specific gene (also called TSA gene); however, Arai et al54 found that groEL gene has strong conservation rather than the 56 kDa gene which is also distinct from other Rickettsia species. In India, 56 kDa gene has been used to study strain prevalence rate, Karp-like2252555657 Kato-like56 strains are more prevalent and the least prevalent are Gilliam-like22525658and TA678 strains2252. New genotypes denoted as IHS has been identified in Shimla5556.

Recently, Batty et al59 have sequenced the complete genome of six O. tsutsugamushi strains, namely Karp (including UT76 and UT176), Kato, Gilliam, TA686, TA763 and TA716 (FPW1038) and compared them with existing reference strains Boryong and Ikeda. Based on the 657 core genes observation, Kato and Ikeda’s strains are more closely related to Karp strains (incl. UT76 and UT176) than TA686 and Gilliam, as reported in 56 kDa tree59. It is also noted that a particular gene or a single gene could not be used for strain identification, i.e. Karp-like and Kato-like. In this study, it was observed that groEL of Karp (LS398548) and Ikeda (NC_010793/AP008981) reference strains shared about 98.5 per cent identity, i.e. only 25 mismatch nucleotides (

Taken together, we report that our sequences belong to O. tsutsugamushi; however, at this time, we were unsure of placing it as a particular strain due to lack of complete genome sequence. Taken together, our observation pointed out only the rural outdoor collected rodents/shrews such as R. rattus, B. bengalensis and S. murinus and their ectoparasites were positive for scrub typhus pathogen, O. tsutsugamushi. We believe that this is the first report for the presence of O. tsutsugamushi in B. bengalensis as well. Contrary to the claim that among the ectoparasites, established vectors trombiculid mites were found to be negative for O. tsutsugamushi in the study areas; however, adult non-trombiculid mites such as Oribatida species, Dermanyssus gallinae, fleas (Xenopsylla astia, X. cheopis, Ctenophalides felis, Ctenophalides sp.) and ticks (Rhipicephalus sanguineus, R. haemaphysaloides) were found to be positive for O. tsutsugamushi and this is a first report globally. Although there are no reports on these ectoparasites as vectors for scrub typhus, there is a possibility that these ectoparasites may act as a vector in the transmission of scrub typhus. Further research will be carried out to assess whether these ectoparasites act as vectors in the transmission of scrub typhus.

The O. tsutsugamushi groEL gene sequence similarity was much pronounced between the host and their ectoparasites, indicating the transmission of the pathogen to the host or vice versa, which shows the transmission of scrub typhus pathogen between host and its ectoparasites in this region. Since both rodents/shrews and the ectoparasite vectors live near the human niche, there is a possibility of risk for the transmission of scrub typhus. Thus, it is suggested to undertake routine host/vector/pathogen surveillance to identify hotspots to implement appropriate preventive measures. However, it needs further validation and confirmation, followed by laboratory experiments to prove the potential transmission of the acquired infection by the newer ectoparasites. The outcome of this study gains much public health importance to design appropriate preparedness for control measures and to sensitize medical practitioners for early diagnosis and treatment to reduce mortality associated with scrub typhus.