Determination of Scrub typhus Suggests a New Epidemic Focus in the Anhui Province of China

In 2007, 19 cases of a scrub typhus epidemic occurred within a week at a sports school in Mingguang County, Anhui Province, where no previous incidence of this mite borne disease had been reported. Sero-surveillance in 2009 indicated that 10 of the 100 school students possessed anti-Orientia tsutsugamushi antibodies. From 2009 to 2013, 60 small animals and 2250 mites were collected in the vicinity of the school. 5 of the Apodemus agrarius samples and 1 group of Leptotrombidium linhuaikongense tested positive via PCR for O. tsutsugamushi. Two strains of O. tsutsugamushi were identified by injecting Kun Ming (KM) mice peritoneally with the organs of either Apodemus agrarius or Leptotrombidium linhuaikongense. Apart from sharing 98% homology with the O. tsutsugamushi Yongworl strain, genes encoding the membrane protein from the two O. tsutsugamushi isolates shared >99% sequence homology with each other, reflecting the consistency of the pathogen in both the vector and the host. In addition, we also characterized a chronic scrub typhus infection in a local patient. The membrane protein gene fragment from the patient’s blood shared 99% homology with O. tsutsugamushi Gilliam strain, suggesting that more than one O. tsutsugamushi strain is present at this location.

in Fuyang city, in the northwestern part of Anhui Province 13 . As an investigation into the natural foci of scrub typhus in this province hasn't previously been conducted, this study was designed to investigate scrub typhus transmission in the vicinity of the sports school at which the epidemic of scrub typhus was reported in 2007. A focused investigation designed to identify O. tsutsugamushi genotypes, vector species and host animals present in Mingguang 13 was conducted in order to highlight the features of scrub typhus at this location and to implement measures for preventing further dissemination of the disease.

Results
Serological survey. A scrub typhus epidemic consisting of 19 cases broke out in 2007 in a sports school in the Anhui province of northern China. A serological surveillance performed in 2009, using a colloidal gold reagent, indicated that 10 out of the 100 school students possessed anti-O. tsutsugamushi antibodies. On retrospection, five of the students who tested positive recalled that they had caught fevers or coughs in September 2008, the symptoms of which were alleviated after receiving oral doxycycline provided by a school doctor. The other 5 students recalled experiencing symptoms of a cold and received no treatment, suggesting the existence of sub-clinical infection at this location.

Determination of host animals and vector mites. Small animals were collected annually from 2009
to 2013, in October using mouse cages positioned for a week in the vicinity of the sports school. In total 60 small animals were captured; 38 Apodemus agrarius, 16 R. norvegicus, 2 Microtus fortis and 4 Neomys fodiens, indicating that Apodemus agrarius was the dominant species, accounting for 63.3% of the total small animals captured (Table 1). 2250 mites were collected from the ears of the trapped small animals, consisting of 1980 Leptotrombidium linhuaikongense and 270 Odontacarus majesticus, indicating that Leptotrombidium linhuaikongense was the dominant mite species. The rate at which the small animals above carried mites are as follows: 60.5% (23/38) Apodemus agrarius, 31.3% (5/16) R. norvegicus, and 0 Microtus fortis or Neomys fodiens. The total rate of carrying mites was 46.7% (Table 1). Forty small animals including 25 Apodemus agrarius, 11 R. norvegicus, 2 Microtus fortis and 2 Neomys fodiens were captured in 2009. However, due to the rodent control strategies carried out by the sports school since 2010, annually, only 10, 5, 5 and 0 small animals were captured thereafter.

Isolation of O. tsutsugamushi.
The 40 small animals trapped in 2009 were divided into 7 groups according to habitat and species; 3 groups of Apodemus agrarius, 2 groups of Rattus norvegicus, 1 group of Microtus fortis and 1 group of Neomys fodiens. The organs from these animals were dissected, triturated and pooled to yield one sample per group. The chiggers collected in 2009 and 2010 were pooled, then divided by species into one group of Leptotrombidium linhuaikongense and one group of Odontacarus majesticus, which were also triturated. KM mice were inoculated with these samples, and mice presenting symptoms of infection after two weeks were sacrificed. Liquid was scraped from the abdominal walls of infected mice, and the presence of O. tsutsugamushi particles was assessed by light microscopy. Mice inoculated by 1 specimen from Apodemus agrarius and 1 specimen of Leptotrombidium linhuaikongense presented typical features of O. tsutsugamushi infection, and this was confirmed via abdominal wall scrapings which revealed a O. tsutsugamushi infection (Fig. 1).

PCR detection of O. tsutsugamushi.
The organs of all 60 small animals and five groups of 20 chiggers (4 groups of Leptotrombidium linhuaikongense and 1 group of Odontacarus majesticus) and the organs of the infected mice, as described above, were triturated and DNA was extracted. O. tsutsugamushi infection was detected by PCR amplification of a specific 730 bp gene fragment of O. tsutsugamushi. The rate of detection of O. tsutsugamushi in wild small animals was 8.3% ( Fig. 2A). The specific gene fragments of O. tsutsugamushi were also detected from the organs of inoculated mice (Fig. 2B) When patient blood genomic DNA was used as a template, the 730 bp target gene was amplified by PCR and analyzed by agarose gel electrophoresis (Fig. 2C). However, after receiving doxycycline treatment for a month, the specific gene fragment of O. tsutsugamushi was absent from the patient's blood (Fig. 2C).
Sequence analysis and comparison. The 56 kDa protein sequences of the O. tsutsugamushi strains isolated from the Apodemus agrarius and Leptotrombidium linhuaikongense were designated sanjie and sanjie2, submitted to Genbank and assigned the serial number KM095135 and KM115577, respectively. The 56 kDa protein sequences of the O. tsutsugamushi isolates from the host animals and vector mites shared > 99.8% homology, reflecting the consistency of pathogens between host and vector, and shared 98% homology with O. tsutsugamushi Yongworl strain (an isolate from Korea). Phylogenetic tree analysis (Fig. 3) also indicated that local strains and Korea young-worl strains belong to the same branch. The sequence of the specific gene fragment amplified from the patient's blood was named sanjie3 and assigned the serial number KM115578. It shared 100% homology with O. tsutsugamushi Hefei strain and 99% homology to the Gilliam strain.

Discussion
Since the appearance of scrub typhus in Mingguang, neither the epidemic foci of this infection nor the pathogen isolation in this area has been previously reported. An earlier report of another epidemic in Fuyang, a different area in the Anhui province ( Fig. 4) merely mentioned the characteristic of the patients. This study conducts an epidemiological investigation of the natural focus of scrub typhus in the vicinity of a sports school that experienced an outbreak of scrub typhus in 2007. A 2009 sero-survey determined that 10% of students at this school possessed anti-O. tsutsugamushi antibodies. This rate is lower than that amongst the other residents of Mingguang, which was as high as 71.50% (143/200) 14 . This difference may be a reflection of the length of time spent by the patients in this area. Students of the sports school were mainly from different regions of China, while local residents were most likely to have resided in the area for their entire lives. In addition, students may have been less exposed than local residents to the vectors or vector hosts. Nonetheless, the high positive rate of anti-O. tsutsugamushi antibodies indicates a high risk of exposure to O. tsutsugamushi infection. Personal protection was recommended for those who frequently visit the native habitats of small animals and chiggers in the field. The small wild animals trapped in autumn included Apodemus agrarius, R. norvegicus, Microtus fortis and Neomys fodiens with Apodemus agrarius accounting for 63.3% of the total small animals trapped. Although the number of the trapped small animals was small, likely due to the prevention strategies we recommended to reduce the risk of infection at this site, the pathogen was nonetheless detected in the dominant rodent species, Apodemus agrarius, indicating that it might be the major host for O. tsutsugamushi at the location, similar to reports from the neighboring provinces of Jiangsu and Shandong.   (1) and Leptotrombidium linhuaikongense (2); C: health mice used as control. (C) PCR products from the patient with scrub typhus M: 1 kb marker; 1: PCR products from the blood of a patient with chronic infection in; 2: Negative amplification from the blood of the patient after symptomatic treatment; C: Blood from healthy adults used as control.
Scientific RepoRts | 6:20737 | DOI: 10.1038/srep20737 The two species of mites collected from the ears of trapped small animals were identified as Leptotrombidium linhuaikongense and Odontacarus majesticus. Leptotrombidium linhuaikongense was found in the Linhuaigang district of Huoshan County in Anhui province for the first time in 1959 15 , which is over two hundred miles away from the county of Mingguang. Although the specific gene fragment of O. tsutsugamushi had been amplified from this species of chigger previously 16 , we found no report that Leptotrombidium linhuaikongense is a vector of O. tsutsugamushi. In China several chiggers have been confirmed to be vectors of O. tsutsugamushi, including Leptotrombidium deliense, Leptotrombidium rubellum, Leptotrombidium gaohuense, Leptotrombidium insulare, Jishou Leptotrombidium and Leptotrombidium scutellare 17 . In this investigation, Leptotrombidium linhuaikongense  was the dominant species in collected mites (88%). The specific gene segment of O. tsutsugamushi was amplified and a new O. tsutsugamushi strain was isolated, indicating that this species of mite is likely to be a new vector of scrub typhus. A specific gene fragment of O. tsutsugamushi was also amplified from the blood of a patient with a chronic O. tsutsugamushi infection, residing near the sports school. Based on the patient's account, we can infer that patient remained infected for about 9 months before symptomatic treatment. After correct diagnosis of scrub typhus, he took doxycycline 0.2 g a day for a month, and the symptoms were alleviated and eventually resolved. PCR detection for a specific gene fragment of O. tsutsugamushi in the patient's blood confirmed that the infection was then cleared. This is the first report of a chronic scrub typhus infection in China.
In fact, chronic infection of O. tsutsugamushi is a common phenomenon among animals, with some kinds of animals carrying the pathogen for a long period without any symptoms 18,19 . Persistence of the pathogen in the human body is rarely reported, mostly because medical staff rarely track the persistence of pathogens after resolution of clinical symptoms. Moon-Hyun Chung and others issued a detailed report in 2013 20 where they found that a specific DNA fragment could still be detected in the patient's blood four months after infection. O. tsutsugamushi was isolated from the blood of a patient 18 months after infection following symptomatic treatment, and in the absence of typical clinical symptoms of scrub typhus. In our study, the patient had never received the symptomatic treatment before correct diagnosis, which may explain the detection of O. tsutsugamushi DNA in the blood nine months after infection. The anti-O. tsutsugamushi antibodies in this patient's serum could not be detected by the colloidal gold reagent developed by our group, while a more sensitive ELISA determined the anti-O. tsutsugamushi IgG antibody titer of the patient's serum to be 1:400. This result echoes a previous report that the anti-O. tsutsugamushi antibody titer in the serum of patients would be significantly reduced 4 to 8 months after infection, in comparison with that during the acute period 18 . In this case disease was often reported in October or November, in accordance with cases reported north of the Yangtze River in China, which all occurred in autumn or winter. The Gilliam and Karp genotypes of O. tsutsugamushi are thought to be most prevelet in south of Yangtze River 21 , while Kawasaki and Gilliam types were mainly found in north of the Yangtze River 22,23 . In this study, sequences of the gene encoding the 56 kDa outer membrane protein of the two O. tsutsugamushi isolates from wild small animals and vector mites shared > 99.8% homology, reflecting the consistency of pathogens between host and vector. They also shared 99% homology with O. tsutsugamushi Yongworl strain, an isolate from Korea which had previously never been found elsewhere in China. Phylogenetic tree analysis also indicated that the local strains and Korea young-worl strains belong to the same branch. Meanwhile, the sequence of the specific gene fragment of O. tsutsugamushi from the patient's blood shared 100% homology with that from the first patient in the city of Hefei, and is closely related to the Gilliam strain.
These results suggest that more than one strain of O. tsutsugamushi is circulating in this area. Our conclusions are limited by the number of blood samples we could obtain from local patients diagnosed with scrub typhus. During the outbreak of scrub typhus in 2007, only patients' sera were retained, while the clot was discarded due to lack of prior experience. Hence the genotype of O. tsutsugamushi could not be determined at that time. Although cases of scrub typhus have been occasionally reported in recent years, patient specimens were not obtained in a timely fashion as it was not required by law to officially report scrub typhus in China. In addition, the disease is often misdiagnosed due to sharing similar symptoms with other febrile diseases such as the common cold. We cannot determine the prevalence of O. tsutsugamushi strains at this site through analysis of only one patient. Thus, further work will be required to continue this line of research.
In conclusions, this is the first report identifying vectors and animal hosts of the O. tsutsugamushi genotypes in Mingguang. Briefly, Apodemus agrarius and Leptotrombidium linhuaikongense are likely to be the most prevalent host and vector of scrub typhus in this area. Gilliam and Yongworl-like strains of O. tsutsugamushi were found. Chronic scrub typhus infection was also reported for the first time in a Chinese patient. These studies highlight the features of scrub typhus at the location and may provide useful information for further understanding the distribution of scrub typhus in China, facilitating targeted prevention and treatment of the disease.

Methods
Ethics Statement. The study was conducted on private lands, in the sports school and in the wild field.
Prior permission was obtained from the president of the sports school, Mr. Huadong Huang, and the owners of the private lands named Mingqiang Li and Xing Xu. In China, no ethical approval is needed for sampling rodents outside protected areas and the field study didn't involve any endangered or protected species.
All animal experiments were approved by the Ethics Committee of the Research Institute for Medicine, Nanjing Command. These methods were carried out in accordance with the approved guidelines. Human participation in this study was approved by the Ethics Committee of Research Institute for Medicine of Nanjing Command and the people provided written informed consent.
Blood sample collection for serological surveys. In 2009, 100 sera samples of students, over the age of 18, attending the sports school at which scrub typhus was detected in 2007 were collected. The O. tsutsugamushi specific antibody content of participant sera was detected using colloidal gold diagnostic reagents, as previously described 24 . Blood sample collection from a patient chronically infected with O. tsutsugamushi. In May 2013, a young male presented with enlargement of neck, armpit and inguinal lymph nodes of unknown etiology. The patient described in retrospection that he had a fever over 39 °C in September 2012 which was accompanied by a rash all over the body. The fever persisted for 2 weeks despite administration of cephalosporin antibiotics, then resolved. He recalled that he was bitten by an insect which left an eschar-like scar near his navel. However, the doctor ignored this feature at that time. This patient had subsequently experienced repeated mild fever accompanied by lymph node enlargement, and had sought medical attention, but was not properly diagnosed. As this patient was resident in the vicinity of the school at which the 2007 scrub typhus epidemic was detected, his blood sample was eventually sent to our laboratory for analysis. Field Sample collection. Small animals were collected in October, annually, from 2009 to 2013. Mouse cages were used to capture mice for a week in locations including houses, vegetable fields, peanut fields, grasslands, ridges and other areas where small animals such as rodents were often seen around or in the sports school. The cages were placed at nightfall and retrieved at dawn and then the trapped rodent species were identified and the dominant species was determined. The mites present inside the ears of each rodent were collected, counted, and the species were identified by microscopy and the dominant mite specie was determined. The small animals were anesthetized by ether and the organs, including the livers, spleens and kidneys, were dissected under sterile conditions and reserved for PCR detection and pathogen isolation. The remains of each rodent specimen were stored at − 80 °C. The mite specimens were used for PCR detection and pathogen isolation or stored in 70% ethanol.

Isolation of O. tsutsugamushi.
The 40 small animals trapped in 2009 were divided into 7 groups according to habitat and species; 3 groups of Apodemus agrarius, 2 groups of Rattus norvegicus, 1 group of Microtus fortis and 1 group of Neomys fodiens with 2~3 small animals per group. The organs of these animals were acquired as previously described, triturated and suspended in SPG sucrose phosphate buffer to a final concentration of 10%, and pooled to yield one sample per group. The chiggers collected in 2009 and 2010 were pooled, then divided by species into one group of 20 Leptotrombidium linhuaikongense and one group of 20 Odontacarus majesticus, which were also triturated. 10 mice of the Kunming (KM) species for each group (purchased from Academy of Military Medical Science) weighing 15 to 18 g each were intra-peritoneally injected with 0.5 ml of one sample. The inoculated mice were observed carefully twice daily. Two weeks after inoculation, mice presenting symptoms such as weakness, trembling and erect hair, were sacrificed after about two weeks. Liquid was scraped from the abdominal walls of infected mice, then fixed and stained with Giemsa stain, and the presence of O. tsutsugamushi particles was assessed by light microscopy.
PCR detection of O. tsutsugamushi from field trapped small animals and mites, the inoculated mice and patient's blood. The organs (including liver, spleen and kidney) of 60 small animals and five groups of 20 chiggers (4 groups of Leptotrombidium linhuaikongense and 1 group of Odontacarus majesticus) were triturated and DNA was extracted using the Invitrogen DNeasy Blood & Tissue kit, following the manufacturer's instructions. DNA were extracted from organs of the inoculated mice presenting obvious infection of O. tsutsugamushi and from the patient's blood using the same Invitrogen kits. PCR was performed with the following primers designed according to the gene sequence of the 56 kDa outer membrane protein of O. tsutsugamushi standard strains; P1: 5′ -gagcagagataggtgttatgtacc-3′ , P2: 5′ -tatacccatcaaaaagatctctgagc-3′ 25  The reaction ended with a final step of 7 min at 72 °C. PCR products were analyzed on 1% agarose gel, and the sizes of the PCR products were determined against a standard molecular weight marker (Fermentas MBI Co., Lithuania, and TaKaRa Co.).