Genotypic characterization of Orientia tsutsugamushi from patients in two geographical locations in Sri Lanka

Scrub typhus (ST), or tsutsugamushi disease, is an acute febrile illness in humans caused by infection with Orientia tsutsugamushi (OT) following a bite of an infected mite vector of the genus Leptotrombidium [1]. Scrub typhus is endemic in the Asia-Pacific region, extending from Afghanistan to China, Korea, the islands of the western Pacific and Indian Oceans, and northern Australia [2, 3]. This endemic region is often referred to as the tsutsugamushi triangle, and hosts approximately 1 billion people [4]. The vectors can be found in a variety of ecological conditions, from the mountainous regions of northern India to the tropical climates of the Malay Peninsula and Indonesia [2]. Transovarial transmission of OT within vectors appears to be essential to maintenance of the agent in nature; thus, the mite serves as both the vector and the reservoir [2]. Transmission of the etiologic agent to vertebrate hosts occurs during feeding of the larval or “chigger” stage of mites [2]. While Orientia is vertically maintained in Leptotrombidium mite populations, it may be transmitted horizontally from mites to vertebrate hosts [2]. The transmission to humans is incidental. Currently, there is no vaccine against ST [5].

The disease mimics several other tropical febrile illnesses, and can vary from mild to fatal disease, with reported mortality rates of 35%–50% during the pre-antibiotic era [6, 7] to 7–9% currently, in different geographical regions [5]. Although detection of eschars assists early diagnosis, the occurrence of this sign is highly variable and depends on the host, geographical region, and possibly the bacterial strain [8]. ST became more familiar during World War II, when soldiers deployed to endemic regions were affected in great numbers [2]. Research since World War II has highlighted dramatic antigenic variation among strains of OT [2] Although great inter-strain variability in virulence has been shown in mouse models, it is not clear whether virulence for mice can be directly applicable to humans [2]. However, the antigenic variation may result in a spectrum of illness from very mild to disease that is often fatal when untreated [2]. To date, 20 antigenically distinct strains of OT have been reported, including the initially serologically distinguished prototypic strains Karp, Gilliam, and Kato [9]. The introduction of chloramphenicol and, later, the tetracyclines dramatically eliminated mortality among cases where treatment was started early [10‐12]. Although relapses and reinfections occur [12], infection is generally responsive to treatment with antibiotics such as doxycycline and azithromycin, even when antigenically diverse strains are involved [2].

The documented history of scrub typhus in Sri Lanka dates back to the Second World War [5]. Similar to other countries in the tsutsugamushi triangle where scrub typhus is endemic, Sri Lanka also experiences a variable burden due to the illness. The incidence of the disease seems to be based on human activity, and ecological and climatic factors. Although OT in Sri Lanka is confirmed with IFA-based diagnostic techniques, the genotypic characterization of infecting OT strains is unknown. Such genotypic characterization of OT is important to establish effective diagnostics and to develop effective vaccines.

The current study was designed to identify the strains of OT causing clinical illness in two areas along the western coastal belt of Sri Lanka. Both areas (Fig. 1), which are approximately 100 km apart, are endemic for OT. They are likely to have similar ecological characteristics, both are in the wet zone and 5–10 km from the coast.

Paediatric and adult patients who presented with an acute febrile illness to Balapitiya Base Hospital (Location 1) and the Professorial Medical Unit of Colombo North Teaching Hospital (Location 2) in Sri Lanka were examined clinically in order to identify an eschar; the essential inclusion criteria for the study. Consecutive patients who fulfilled the essential criteria (acute fever and an eschar) were recruited for the study. For patients who were more than 18 years of age a punch biopsy of the eschar was performed after obtaining informed written consent. For those who were less than 18 years old, as they were considered not competent for consenting according to the Sri Lankan law, a 3 cc sample of blood was collected after obtaining informed written consent from their guardians. Patients who were considered severely ill, declined to consent, or had received anti-rickettsial antibiotics for more than 24 h were excluded from the study. Relevant demographic, clinical, and laboratory data of all patients were collected into a predesigned data collectionform.

Buffy coat samples stored in RNAlater were centrifuged in 2.0 mL microcentrifuge tubes at 5000 × g for 5 min. The supernatant was removed, and the remaining cell pellet was resuspended in 200 μL of PBS with 200 μL of lysis buffer (QIAGEN). DNA was extracted using the DNeasy Blood and Tissue Kit.

Real time PCR was performed on extracted DNA using primers designed to amplify a segment of the Orientia 47-kD antigen gene (Additional file 1: Table S1). [13] Of the samples positive for OT DNA, nested PCR to amplify a portion of the 56-kD protein gene was performed as previously described (Additional file 1: Table S1). [14] PCR amplicons were cloned using chemically competent E. coli cells and the TOPO TA Cloning Kit (Life Technologies). Plasmid DNA was purified with the PureLink Plasmid Miniprep Kit (Life Technologies) and sequenced using a 3130xl Genetic Analyzer (Life Technologies).

The partial 56-kD sequences were trimmed of primer sequences using EditSeq of the Lasergene Core Suite software version 12.0 (DNASTAR, Madison, WI) and then identified by comparing them with sequences available in GenBank via BLAST database search (http,//blast.ncbi.nlm.nih.gov). Phylogenetic analyses of the study and reference sequences were performed using MegAlign of the Lasergene Core Suite (DNASTAR). Sequence alignment was performed using CLUSTAL W, and a phylogenetic tree was constructed with 1,000 bootstrap replications using the neighbor-joining method. For residue conservation analysis, nucleotide sequences were translated from open reading frames of DNA sequences using EditSeq. These nucleotide sequences were then aligned and analyzed using MegAlign.

Originally, three distinctive antigenic prototypes of OT (Karp, Kato, and Gilliam) were described using serologic testing (via complement fixation) [15]. The serologic diversity is based largely on the 56-kDa type specific antigen (TSA), [16]. Later, over 20 additional strains of OT were found [9] by indirect immunofluorescence assays (IFA) using strain- or type-specific monoclonal antibodies or hyperimmune sera which recognize the 56-kDa TSA and by the genetic sequences of this genetically diverse protein. Recent efforts to catalog the strains present in various locations have identified novel subtypes. Multi-locus sequence typing (MLST) analysis of Thai isolates has shown a very high rate of recombination, which along with gene duplication and horizontal gene transfer accounts for the genetic diversity among OT isolates in various locations of Thailand [17, 18].

An increased prevalence of Karp and Karp-like strains has been reported in the Southeast Asian countries, including Malaysia (56%), Thailand (65%), Myanmar (46%), and Philippines (78%) [19‐21], Boryong is the predominant strain reported from Korea [22]. However, several studies in India have reported a high prevalence of Kato-like organisms (65%) and the majority have sequence similarity with OT isolates from Korea, Thaiwan, Vietnam and Cambodia [23]. Around 35% were Karp-like, and the rest were related to Kawasaki and Gilliam types.

A 2011 study in Korea comparing the clinical severity between Boryong and Karp genotypes suggested that eschars and rashes were found in 97% and 94% of the patients infected with the Boryong cluster compared to 73.7% and 68.4% of the patients infected with the Karp cluster, respectively, suggesting clinical variation among infections with different serotypes of OT [24]. In a previous study conducted in the same geographical region, we found a high eschar detection rate (89%) that is in keeping with the above findings [25].

Since this study analyzed only 16 positive samples, it is difficult to arrive at conclusions on the complete composition of OT genotypes in Sri Lanka. However, since Sri Lanka has diverse ecological and environmental conditions within the matter of 2–3 h travel, a larger scale study that covers all ecological zones could contribute to understanding of the complex diversity of OT infection and its transmission within the country.