To identify the vector species for Shimokoshi type Orientia tsutsugamushi, a survey of larval trombiculid mites was conducted in Yamagata Prefecture, Japan from April to May 2012. In all, 2889 larval trombiculid mites were obtained from 21 Apodemus speciosus rodent hosts, 2600 of which were morphologically classified into eight species in three genera. After screening of O. tsutsugamushi DNA in individual larval trombiculid mites using real-time PCR targeting the 16S ribosomal RNA gene, serotype-specific nested PCRs targeting the 56 kDa protein gene were performed, followed by sequencing analysis. As a result, Shimokoshi type O. tsutsugamushi DNA was identified from 3 (1.9%) of 157 Leptotrombidium palpale. This is the first study to identify Shimokoshi type O. tsutsugamushi DNA in L. palpale. The results indicate that L. palpale is a possible vector for Shimokoshi type O. tsutsugamushi.
- G. saduski
indirect immunofluorescence assay
- E. ichikawa
Scrub typhus, also known as tsutsugamushi disease, is an acute febrile disease that is characterized by fever, rash, and eschar. The disease is transmitted to humans by the bite of larval trombiculid mites harboring O. tsutsugamushi and is endemic in the Asia-Pacific region . In Japan, scrub typhus is the most common rickettsiosis, more than 300 cases being reported each year . O. tsutsugamushi strains found in humans in Japan are classified into six serotypes: Gilliam, Karp, Kato, Kawasaki, Kuroki and Shimokoshi . Currently, Gilliam, Karp, Kawasaki and Kuroki are the major scrub typhus serotypes associated with human infections [3, 4]. Kato type is rarely reported  and Shimokoshi type was first reported in 1984 , only a few cases having been reported thereafter [6-8]. These serotypes are genetically distinguishable by nested PCRs targeting the O. tsutsugamushi 56 kDa protein gene [6, 9].
It is presumed that human scrub typhus is primarily transmitted by mites of the genus Leptotrombidium . In Japan, L. pallidum, L. scutellare and L. akamushi are believed to transmit Gilliam and Karp types, Kawasaki and Kuroki types and Kato type, respectively . However, the vector of the Shimokoshi type is yet to be identified; this failure acts as a bottleneck in the clarification of the epidemiology of Shimokoshi type scrub typhus in Japan.
The Tohoku region, which includes Yamagata Prefecture, is located in the northern part of Japan's main island of Honshu and is known to be endemic for Gilliam and Karp types . Recent studies have shown that 90% of cases of Karp type scrub typhus are caused by the JP-1 type, which is a Karp genotype [11, 12]. As for the Shimokoshi type, only eight cases have been reported to date [6-8]. In the present study, we used larval trombiculid mites to conduct a survey in Yamagata Prefecture, Japan to investigate potential vectors for Shimokoshi type O. tsutsugamushi.
MATERIALS AND METHODS
Investigation of larval trombiculid mites
Collection of larval trombiculid mites from field rodents
To collect larval trombiculid mites, Sherman live traps were used to capture field rodents from April to May 2012 in Asahi-machi, Yamagata Prefecture, Japan, the site of a recent case of Shimokoshi type scrub typhus (in May 2011) . After the animals had been killed, serum and spleen were collected aseptically and kept at −80°C until use. The carcass of each rodent was hung over a beaker filled with water to collect larval trombiculid mites and the collected mites washed three times with sterilized distilled water. Each larval trombiculid mite was individually placed in a well of a 24-well slide glass and the lower abdomen of each mite punctured with the acuminated tip of a 0.02 mm diameter disposable glass rod under a stereoscopic microscope. Subsequently, their body fluids were harvested in 0.2 mL PCR tubes containing 6 μL of sterilized distilled water and kept at −30°C after heating (100°C, 10 mins). After harvest of the fluids, the remaining mite shells were mounted in gum-chloral solution and subjected to morphological identification of species under a differential interference contrast microscope. Mite samples were labeled so that; for example, R08-m133 represented mite number 133 collected from rodent number 8.
The field rodents were captured with the permission of the Yamagata Prefectural authorities. All animal experiments were approved by the Yamagata Prefectural Institute of Public Health, Institutional Animal Care and Use Committee and conducted according to their guidelines.
Orientia tsutsugamushi-specific real-time polymerase chain reaction
To screen for O. tsutsugamushi infection in the larval trombiculid mites, a real-time PCR system based on the 16S rRNA gene was developed using the primers Ot16S-1008F: 5′-CGTGTCGTGAGATGTTGGGTTA-3′ and Ot16S-1073R: 5′-ACCCGCTGGCAAATAAGAATAAG-3′ and the TaqMan probe Ot16S-1032P: 5′-VIC-TCCCGCAACGAGC-MGB-3′. Through comparison with the 16S rRNA gene sequences of Gilliam (GenBank accession no. D38622), Karp (D38623), Kato (D38624), Kawasaki (D38625), Kuroki (D38626) and Shimokoshi types (D38627) , these primers and probe were designed to specifically detect all known O. tsutsugamushi types. The reactions were performed using TaqMan Fast Advanced Master Mix on a 7500 Fast Real-Time PCR System (Applied Biosystems, Weiterstadt, Germany) with 1.0 μL template, 500 nM forward and reverse primers, and 200 nM TaqMan probe in a total reaction volume of 10 μL. Amplification conditions consisted of initial denaturation at 95°C for 20 sec, followed by 38 two-step cycles of 95°C for 3 sec and 60°C for 30 sec. In this assay, samples were considered to be positive when the δRn value at 38 cycles was above 0.2.
Serotype-specific nested polymerase chain reactions and sequencing analysis
To determine serotypes of O. tsutsugamushi, serotype-specific nested PCRs targeting the O. tsutsugamushi 56 kDa protein gene were conducted as described previously . Shimokoshi type-specific primers were designed based on the same gene (Table 1) and three different combinations of nested PCRs performed as follows: (i) first PCR with Shi193F-Shi1084R (95°C for 4 mins and 40 cycles of 95°C for 30 sec, 55°C for 30 sec and 72°C for 1 mins) and second PCR with Shi402F-Shi942R (95°C for 4 mins and 40 cycles of 95°C for 30 sec, 60°C for 45 sec and 72°C for 1 mins); (ii) first PCR with Shi193F-Shi2233R (94°C for 5 mins and 35 cycles of 95°C for 30 sec, 55°C for 30 sec and 72°C for 2 mins) and second PCR with Shi402F-Shi2206R (94°C for 5 mins and 35 cycles of 95°C for 30 sec, 55°C for 30 sec and 72°C for 2 mins); and (iii) first PCR with Shi193F-Shi2233R and second PCR with Shi402F-Shi942R. The reactions were performed using GoTaq Green Master Mix (Promega, Madison, WI, USA) with 50 nM forward and reverse primers. The template volumes used in the first and second reactions were 0.5 μL (50 μL reaction volume) and 2.5 μL (25 μL reaction volume), respectively. The PCR products were subjected to direct sequencing and the resulting sequences compared with the sequences in the database using the basic local alignment search tool (BLAST) search program available at the DNA Data Bank of Japan website (http://blast.ddbj.nig.ac.jp/blast/blastn?lang=ja).
|Primer name||Sequence (5′–3′)||Purpose|
Investigation of field rodents
Infection status of Orientia tsutsugamushi
Field rodents infested with larval trombiculid mites proven to be positive for O. tsutsugamushi infection by nested PCR and sequencing analysis were further analyzed in the following experiments.
First, indirect immunofluorescence assays were carried out to detect O. tsutsugamushi-specific serum antibodies in the rodents. Standard strains of the six O. tsutsugamushi serotypes (Gilliam, Karp, Kato, Kawasaki, Kuroki and Shimokoshi) were cultured in L929 cells. One milliliter of the cultured cells was centrifuged at 15000 g for 15 mins. The resultant pellets were washed once in PBS and mixed with 50 μL of PBS supplemented with 0.1% (v/v) formalin and 0.3% (v/v) FBS. Each antigen was spotted on a 24-well slide glass and fixed with acetone for 15 mins. Serum samples from the field rodents were serially diluted in PBS to 1:40, 1:80, 1:160, 1:320 and 1:640. The diluted sera were incubated with antigens for 40 mins at 37°C. The slide glasses were washed three times in PBS, then reacted with 100-fold diluted FITC-conjugated mouse IgG-heavy and light chain cross-absorbed antibody (Bethyl, Montgomery, TX, USA) for 40 mins at 37°C. After washing three times with PBS, specimens were observed under a fluorescence microscope. In this assay, samples were considered to be positive when the antibody titer was ≥ 1:80.
Second, serotype-specific nested PCRs were carried out to detect O. tsutsugamushi DNA in the rodents. DNA was extracted from the spleen of the rodents using a QIAamp DNA Mini Kit (Qiagen, MD, Gaithersburg, USA) according to the manufacturer's instructions. The PCR reaction was performed as described above except that the template volume in the first reaction was 5 μL.
Analysis of larval trombiculid mites
In all, 2889 larval trombiculid mites were obtained from 21 Apodemus speciosus rodents, 289 of which were excluded from the present study because of structural damage. A final total of 2600 mites were morphologically classified into eight species in three genera. The predominant species was L. pallidum (n = 1420, 54.6%), followed by L. kitasatoi (421, 16.2%), L. intermedium (253, 9.7%), L. fuji (248, 9.5%), L. palpale(157, 6.0%), G. saduski (96, 3.7%), and L. owuense (4, 0.2%). Only one E. ichikawa mite was detected.
Of the 2600 larval trombiculid mites tested, real-time PCR detected 22 O. tsutsugamushi-positive mites (Table 2). Threshold Ct values (0.2) ranged between 28.7 and 38.0. The positive rate was highest for G. saduski (4/96, 4.2%), followed by L. palpale(4/157, 2.5%), L. fuji (2/248, 0.8%), L. pallidum (10/1420, 0.7%), L. intermedium (1/253, 0.4%) and L. kitasatoi (1/421, 0.2%).
|Rodent ID-mite ID||Trombiculid mite species||Real-time PCR, Ct valuea||Nested PCR||Subtypes by sequencing|
|R02-m027||Leptotrombidium pallidum||31.8||−||−||+||−||−||−||−||Karp (JP-1)|
|R08-m056||Leptotrombidium pallidum||37.8||+||−||+||−||−||−||−||Karp (JP-1)|
|R11-m117||Leptotrombidium pallidum||29.0||+||−||+||−||−||−||−||Karp (JP-1)|
|R14-m414||Leptotrombidium pallidum||28.7||+||−||+||−||−||−||−||Karp (JP-1)|
Serotype-specific nested PCRs were performed on 22 samples that had been found positive by real-time PCR (Table 2). Consequently, three samples, R08-m133, R14-m124 and R14-m360, all of which were derived from L. palpale, were found to be positive for Shimokoshi type O. tsutsugamushi by serotype-specific PCR. In addition, four samples, R02-m027, R08-m056, R11-m117 and R14-m414, derived from L. pallidum were positive for Karp type O. tsutsugamushi. The nucleotide sequences of these seven PCR products were determined (GenBank accession no. AB751254-60). A BLAST search revealed that three sequences from L. palpale were identical with or differed by one nucleotide from that of the Shimokoshi type reference strain (accession no. M63381); the deduced amino acid sequences showed 100% identity. Four sequences from L. pallidum were 100% identical to that of the JP-1 reference strain (accession no. AB617591). The overall prevalence of Shimokoshi type O. tsutsugamushi in L. palpale was 1.9% (3/157), whereas that of JP-1 type in L. pallidum was 0.3% (4/1420).
Differential interference contrast microscopic images of L. palpale(R14-m360) and L. pallidum (R14-m414) with O. tsutsugamushi DNA-positive body fluids are shown in Figure 1. All L. palpale mites were characterized and identified by the shape of their scuta, leg III coxal seta position (located at the anterior border), number of dorsal setae and presence of branched ventral palpal tibial setae. All L. pallidum mites were identified by the shape of their scuta and number and shape of dorsal setae.
Analysis of field rodents
Table 3 summarizes the number of larval trombiculid mites found in four A. speciosus rodents (R02, R08, R11 and R14) that were infested with O. tsutsugamushi DNA-positive mites. In addition to L. palpale and L. pallidum, some of which were positive for O. tsutsugamushi, five other mite species (L. kitasatoi, L. intermedium, L. fuji, G. saduski and E. ichikawa) were found on these rodents. Table 4 shows the result of IFA and nested PCRs conducted on these four rodents. Only one sample (R11) was found to be seropositive against both Karp and Shimokoshi types. However, all spleen DNA samples in the nested PCRs were negative.
|Rodent ID||Sex||Weight (g)||No. of larval trombiculid mites|
|L. palliduma||L. kitasatoi||L. intermedium||L. fuji||L. palpaleb||G. saduski||E. ichikawa|
|R08||Male||28.1||199 (1)||9||37||2||8 (1)||1||0|
|R14||Male||58.5||167 (1)||156||65||18||15 (2)||5||1|
|Rodent ID||Serum IgG antibody titer||Nested PCRb|
|R02||< 40||< 40||< 40||< 40||< 40||< 40||−||−||−||−||−||−||−|
|R08||< 40||< 40||< 40||< 40||< 40||< 40||−||−||−||−||−||−||−|
|R11||< 40||80||< 40||< 40||< 40||320||−||−||−||−||−||−||−|
|R14||< 40||< 40||< 40||< 40||< 40||< 40||−||−||−||−||−||−||−|
This is the first study to detect Shimokoshi type O. tsutsugamushi DNA in L. palpale. A previous study reported that, like L. pallidum, L. scutellare and L. akamushi, which have been proved to transmit O. tsutsugamushi to humans, L palpale also bites humans . Furthermore, researchers isolated uncharacterized O. tsutsugamushi-like organisms from L. palpale in Japan between 1960s and 1980s before Shimokoshi type O. tsutsugamushi had been recognized . In this study, we detected Shimokoshi type O. tsutsugamushi DNA from three mites, all of which we identified as L. palpale, whereas other Leptotrombidium species were negative for this type (Table 2). Taken together, these results indicate that it is highly possible that L. palpale is a key reservoir for Shimokoshi type O. tsutsugamushi and thus that it transmits Shimokoshi type scrub typhus to humans. In addition, of nine cases diagnosed with Shimokoshi type scrub typhus in Japan, six occurred in spring and three in autumn [5-8]. L. palpale is mainly found during spring in natural environments [15, 16]. Consistent with this, we collected three L. palpale mites positive for Shimokoshi type O. tsutsugamushi DNA between April and May. The seasonal pattern of activity of these mites also tends to support our contention that L. palpale is a potential vector of Shimokoshi type O. tsutsugamushi.
Karp type O. tsutsugamushi can be divided into two genotypes (JP-1 and JP-2) based on the 56 kDa type-specific antigen protein gene . The vector for JP-2 type is known to be L. pallidum , whereas that for JP-1 type is controversial. In this study, we identified JP-1 type O. tsutsugamushi DNA in L. pallidum (Table 2). Because L. pallidum is the predominant trombiculid mite species in the Tohoku region  and JP-1 type scrub typhus is dominant in this region , it is reasonable to speculate that L. pallidum transmits JP-1 type O. tsutsugamushi to humans. In an earlier study, researchers also isolated JP-1 type O. tsutsugamushi from L. intermedium . However, as this species rarely bites humans , L. pallidum is more likely to play a significant role in disease transmission to humans.
It should be mentioned that this study has the following limitations. First, because we collected larval trombiculid mites from field rodents, we cannot exclude the possibility that O. tsutsugamushi-positive larvae pick rickettsia up from infected rodents and transiently harbor them, but are unable to transmit them to humans. We showed that three of four rodents did not have antibodies against O. tsutsugamushi whereas all of them were negative for O. tsutsugamushi infection on the basis of nested PCR (Table 4). Futhermore, whereas these rodents were parasitized by seven different mite species (Table 3), serotype-specific PCRs showed that all species except L. palpale and L. pallidum were negative for both Shimokoshi and JP-1 types (Tables 2 and 3). These lines of evidence suggest that mites that PCR finds positive for O. tsutsugamushi are infected with rickettsia and did not acquire rickettsia or rickettsial DNA transiently from the rodents.
Second, we developed a real-time PCR assay based on the 16S rRNA gene. Since the primers and probe were designed to anneal to a region well conserved among the different serotypes and even among closely related rickettsial species , it is possible that not only the six known serotypes but also other uncharacterized serotypes of O. tsutsugamushi and/or other related rickettsial species are detected by this real-time PCR. In particular, despite the fact that we found four samples of G. saduski (R01-m044, R16-m090, R16-m102 and R17-m002) and one sample of L. intermedium (R18-m027) to be positive for O. tsutsugamushi by real-time PCR with low Ct values, we found them all to be negative by serotype-specific nested PCR (Table 2), suggesting that these mites may harbor previously uncharacterized O. tsutsugamushi serotypes or related rickettsial species.
Third, serotype-specific nested PCRs were negative in six samples of L. pallidum and one sample of L. palpale(Table 2). Since the assays amplify the gene encoding a 56 kDa outer membrane protein, which is polymorphic , one possibility is that the nucleotide sequences of primer annealing sites were not conserved in those samples. Considering that most of those samples were positive by real-time PCR with high Ct values, it is also possible that serotype-specific nested PCRs were negative due to the difference in sensitivity between nested PCR and real-time PCR.
So far, only a few cases of Shimokoshi type scrub typhus have been reported in Japan [5-8]. This could be because no laboratory diagnostic technique for Shimokoshi type scrub typhus is widely available. Almost all commercial laboratories in Japan only test specimens from patients with suspected scrub typhus for Gilliam, Karp and Kato type antibodies. In addition, the regional institutes for public health, which are in charge of infectious disease diagnosis, seldom test for Shimokoshi type scrub typhus . However, we expect that Shimokoshi type scrub typhus cases may no longer be overlooked because researchers recently reported a nested PCR method for detection of Shimokoshi type O. tsutsugamushi . Based on our results, further investigations, including transmission studies in laboratory animals, are necessary to clarify the role of L. palpale as a vector of Shimokoshi type scrub typhus.
We thank Dr. T. Kobayashi, Department of Internal Medicine National Health Insurance Hospital of Asahi-machi, for his invaluable suggestions. This work was partially supported by a grant-in-aid for the Funding Research Center for Emerging and Re-emerging Infectious Disease from the Ministry of Education, Culture, Sports, Science and Technology of Japan to C.S.
All authors declare they have no conflicts of interests.