Rickettsiae reservoirs among small mammals (Rats, Mice And Shrews) and their Arthropod Vectors in Sri Lanka

Rickettsioses are a group of emerging diseases caused by closely related bacteria. In Sri Lanka, to date, studies have been focused mainly on human subjects. The present study aimed to identify small mammal reservoir hosts and vectors of Rickettsia spp. and Orientia tsutsugamushi in two districts of Sri Lanka. Quantitative-PCR was carried out to detect Rickettsia using citrate synthase gene and Orientia using 47-kD outer membrane protein antigen gene in blood of small rodents and their infested ectoparasites. In both districts ~7.5% blood samples were positive for Rickettsia. Rattus rattus, Bandicota indica and Mus fernandoni were carriers. Three individuals of Suncus murinus, B. indica and Golunda ellioti had only infected ectoparasites. Copies of gltA/100 μL ranged from 133-1.2 × 104 in blood and 197-1.9 × 107 in ectoparasites. Of small mammals with ectoparasites, 43% had Rickettsia positive ectoparasites. Rhipicephalus haemaphysaloides, Ixodes ceylonensis, Haemaphysalis spinigera, Haemaphysalis spp., Stivalius aporus and Xenopsylla cheopis were positive. All study sites except three had infected small mammals or ectoparasites. All samples were negative for O. tsutsugamushi. This is the first study to report Rickettsia spp. in small mammals and their ectoparasites in Sri Lanka. Haemaphysalis spinigera, I. ceylonensis and S. aporus are new records of vectors for Rickettsia. This is also the first report of endemic M. fernandoni as a carrier of Rickettsia and G. ellioti with Rickettsia infected ectoparasites. Though rickettsiosis is not life threatening in most cases, it can lead to severe or fatal disease in vertebrate animals and humans. Hence, the knowledge of the distribution of said pathogen in the reservoirs is essential to control the disease.


INTRODUCTION
Rickettsiae are a group of obligate, intracellular, gramnegative bacteria, including two genera, Rickettsia and Orientia. Rickettsiae cause mild to severe diseases in human and animals collectively known as Rickettsioses (Azad and Beard, 1998). They are transmitted through arthropod vectors, such as ticks, fleas, lice and mites. Genus Rickettsia is classified into two groups, Spotted fever group (SFG) and http://orcid.org/0000-0002-  Typhus group (TG). Orientia tsutsugamushi causes scrub typhus. Rickettsioses have been recognized as an emerging group of diseases in Sri Lanka (Premaratna, 2011;Kularatne et al., 2013). Recent studies have reported predominance of SFG Rickettsioses from Central province and Scrub typhus from Western, North Western, Southern and Northern provinces (Kularatne et al., 2013, Liyanapathirana andThevanesam 2011;Predeepan et al., 2014). Reserch in Sri Lanka up to date have been reported mainly Rickettsioses in human subjects, forcusing on hospital based clinical, epidemiological and serological studies in selected areas (Kularatne et al., 2013, Liyanapathirana andThevanesam, 2011). Only few published data are available on reservoir hosts and vectors of Rickettsiae in Sri Lanka (Nanayakkara et al., 2013;Liyanaarachchi et al., 2012).
Although several arthropod parasites act as vectors for Rickettsia, ticks are the most important in transmission of Rickettsia of SFG. Rhipicephalus sanguineus, and members of the genera Amblyomma and Dermacentor are common vectors of Spotted fever Rickettsia. Murine typhus is transmitted mostly by Xenopsylla cheopis, and Scrub typhus by thromboculid mites (Merhej and Raoult, 2011). Many Rickettsia species are known to vertically transmit in invertebrates, and capable of amplifying within arthropod vectors suggesting that they act as reservoirs of Rickettsia in nature (Azad and Beard, 1998).
In Sri Lanka, rickettsiosis was first recorded in 1937 with 6 Scrub typhus positive patients (Premaratna, 2011). The role of thromboculid mites in transmission of Scrub typhus was studied by Jayawickreme and Nilesin in 1946. First case of Murine typhus was reported in 1938 and its source was confirmed to be rat fleas (Wolf, 1939). At the time it was also reported that typhus-infested rats were transported to the country in Indian vessels (Premaratna, 2011). Later studies reported seroprevalence of Rickettsia among dogs in Kandy, Unawatuna and western slope of central hills (Nanayakkara et al., 2013). SFG Rickettsia was also reported from Amblyomma ticks collected from three wild mammals, pangolin, wild boar, tortoise and Rhipicephalus sanguineus from a dog (Liyanaarachchi et al., 2012). More recently, the presence of Rickettsia was detected in Amblyomma trimaculatum ticks of snake, Boiga forsteni imported to Japan from Sri Lanka (Andoh et al., 2015).
Small mammals, mainly rodent species, for example R. rattus, R. norvegicus, Mus musculus and some wild rodents have been identified as reservoirs or carriers of many Rickettsiae in the world. They also serve as hosts for immature stages of ticks and adult fleas, facilitating the transmission of vector borne diseases (Chen et al., 1998;Hornok et al., 2015). In Sri Lanka, prevalence and distribution of reservoirs of Rickettsiae and species of small mammals and their arthropod parasites involved in transmission of Rickettsiae is not adequately studied. Hence, the objectives of this study were to identify small mammal species and their ectoparasites involved in transmission of Rickettsiae, and their prevalence and distribution in two selected districts in Sri Lanka.

MATERIALS AND METHODS
Small mammals (murine rodents and shrews) were collected from two districts in Sri Lanka, Kurunegala and Kandy, from 2013 to 2014. They were identified using the descriptions given by Phillips (1980). Small mammals were captured using 40 mesh traps placed in each site for four consecutive days. Traps were placed in and around houses, paddy fields, tea and other cultivations. Eight localities were sampled in Kurunegala and 10 in Kandy (Table 1). A 100-300µl sample of blood from saphenous vein and ectoparasites were collected from anesthetized small mammals. Ticks and fleas were identified with the help of taxonomic keys and species descriptions (Kirwan, 1935;Kohls, 1950;Trapido et al., 1964;Seneviratna, 1965;Rajagopalan and Boshell, 1966;Walker et al., 2000;Hopkins and Rothschild, 1953;Mardon, 1981). Ethical clearance for the study was obtained from the Postgraduate Institute of Peradeniya, Sri Lanka.

Sample processing and DNA Extraction
Blood Blood samples were centrifuged at 12,000 rpm for 10 minutes. DNA was extracted from the resulting blood cell pellet using Wizard® Genomic DNA Purification Kit, Promega, USA, according to the manufacturer's protocol for isolation of genomic DNA from blood, with few modifications. Volume of each blood pellet was adjusted to 300 µl by adding PBS. Following modifications were done to the original protocol; Step 3: After adding cell lysis solution centrifugation was done at 14,000 rpm for 1 min; Step 6: After adding Nuclei lysis solution samples were incubated at 80 ºC for 5 min; Step 15: After adding DNA Rehydration solution samples were kept at room temperature overnight to rehydrate the DNA; Step 16: DNA Samples were stored at -20 ºC.

DNA Extraction from Ectoparasites
DNA was extracted using the same extraction kit, according to the manufacturer's protocol for Genomic DNA isolation from animal tissues, with few modifications. Ectoparasites were cut into small pieces using a scalpel blade and transferred into EDTA and nuclei lysis solution. Following modifications were done to the original protocol; Step 3e: After adding Protenase K, tubes were incubated over night at 55 ºC; Step 13: For DNA Rehydration 50 µl of DNA rehydration solution was added and rehydrated overnight at room temperature; Step 14: DNA samples were stored at -20 ºC.

Quantitative PCR assay for Rickettsia
To detect Rickettsia in blood and ectoparasites, seventyfour base pair fragment of citrate synthase gene (gltA) was amplified. Primers used were: Forward-CS-F (5'TCG CAA ATG TTC ACG GTA CTT T 3'), Reverse-CS-R (5' TCG TGC ATT TCT TTC CAT TGT G3') with the probe-CSP (5'FAM-TGC AAT AGC AAG AAC CGT AGG CTG GAT G-TAMsp-3'). The assay specifically amplifies the members of spotted fever and typhus group of Rickettsia. This assay does not produce a positive reaction for the ancestral group Rickettsia, R. bellii nor other members of the order Rickettsiales or any non-Rickettsial bacteria (Stenos et al., 2005). PCR reactions were prepared using 12.5 µl Go taq® Probe qPCR master mix (Promega, USA), 2.5 µl of 2 µM forward and reverse primer and probe, 1 µl of nuclease free water and 4 µl of template. Final volume of the PCR mixture was 25 µl. The thermal profile of the assay, adopted from Stenos et al. (2005) except for the number of cycles, composed of initial holding stage of 50 ºC for 3 min for Pre PCR read and 95 ºC for 5 min for heat activation, 45 cycles of amplification at 95 ºC for 20 sec and 60 ºC for 40 sec and final post PCR read stage of 50 ˚C for 3 min. The reactions were performed in Step one Real Time PCR System (Applied Biosystems, USA) and analyzed using Step One software version 2.2.2. Each sample was run in duplicate with a positive control and two negative controls. An additional third run was performed for samples that gave one positive and one negative result.

Quantitative PCR assay for Orientia tsutsugamushi
To detect Orientia tsutsugamushi in blood of small mammals 118 base pair fragment of 47-kD outer membrane protein antigen/high temperature requirement "A" gene was amplified. Primers used were; Forward -OtsuFP (5'AAC TGA TTT TAT TCA AAC TAA TGC TGC T 3'), Reverse -OtsuRP (5' TAT GCC TGA GTA AGA TAC RTG AAT RGA ATT-3') with Probe 5' FAM-TGG GTA GCT TTG GTG GAC CGA TGT TTA ATC T-TAMsp-3'. The assay specifically amplifies O. tsutsugamushi and does not produce a positive reaction for Rickettsia spp., nor other non-Rickettsial bacteria (Jiang et al., 2005). PCR reactions were prepared using 12.5 µl Go taq® Probe qPCR master mix (Promega, USA), 2 µl of 1.25 µM forward and reverse primers, 2.5 µM probe, 2.5 µl of nuclease free water and 4 µl of template. Final volume of the PCR mixture was 25 µl. Thermal profile of the assay composed of initial holding stage of 60 ºC for 30 s for Pre PCR read and 94 ºC for 5 min for heat activation, 45 cycles of amplification at 94 ºC for 5 s and 60 ºC for 30s and final post PCR read stage of 60 ºC for 30 s. Samples were not run in duplicate since all were negative for Orientia tsutsugamushi. A positive control and two negative controls were used in each run.

Standard curve PCR efficiency and Quantification
To quantify Rickettsia spp. and Orientia spp., and to find out the PCR efficiency, a standard curve was generated using a synthetic double stranded gene fragment, gBlocks® (Integrated DNA Technologies) as

standards (GATATGGGTAAC G G C A T A G T A A C T G A T T T T A T T C A A A C TA AT G C T G C TAT T C ATAT G G G TA G C T T T G G T G G A C C G A T G T T T A A TCTTGAAGGAAAAATTATTGGAATTAATTCT A T T C A T G T A T C T T A C T C A G G C A T A A GTTTTGCTATTCCATCTAATTTTATAAAGCTA T G G G T A T A C C G T C G C A A A T G T T C A C G G TA C T T T T T G C A ATA G C A A G A A C C G T A G G C T G G A T G G C A C A AT G G A A A G A A AT G C A C G A
A G A C C C TGAACAAAAAATCA). It contains the target regions for both Rickettsia spp. and Orientia spp. with 20 additional bases on either side. A ten-fold dilution series of gene fragment was prepared from 3.533x 10 6 to 3.533x10 0 for the standard curve. The last point of the standard curve that had least number of copies (3.533×10 0 ) did not amplify. Hence, the standard curve was prepared with six points, with three replicates for each point, and three negative controls. The approximate quantity was determined using the least Ct values of the positive samples and reading the relevant quantity from the standard curve. For this a common threshold "one" was selected, that go through the linear phase of all the amplification plots of standard curve and the samples. According to the standard curve, the PCR efficiency for Rickettsia assay was 92.1 %, R 2 = 0.994, slope = -3.527, Y intercept = 40.572. According to the standard curve for Orientia, the assay had 87.2 % efficiency, R 2 = 0.999, slope = -3.672 and Y intercept = 43.177.

Kurunegala District
Out of the eight sites in Kurunegala, two sites were negative for Rickettsia; in three sites small mammals were negative but some individuals had positive ectoparasites ( Table 1).
Of the above 120 small mammals, 33 individuals of R. rattus, B. indica and Suncus murinus were infested with external parasites. Of these 33, 11 (33.33%) had Rickettsia positive ectoparasites, 21 were negative for both blood and parasites and one R. rattus was Rickettsia positive and had a negative R. haemaphysaloides larva. Of the 87 small mammals not infested with ectoparasites, 7 were positive for Rickettsia.
Among the ectoparasites carrying Rickettsia were Rhipicephalus haemaphysaloides ticks and X. cheopes fleas. The number of gltA copies in ectoparasites ranged from 197 to 9.1x10 5 . Rhipicephalus haemaphysaloides nymph and a larval pool had high number of Rickettsia, 9.1x10 5 and 4.8x10 5 , respectively. Flea pools of X. cheopes had relatively low number of Rickettsia ranging from 220 to 342 copies (Appendix 2). Most of the (8/11) Rickettsia positive ectoparasites were collected from R. rattus. However, R. haemaphysaloides ticks that had the highest quantity of Rickettsia were from two S. murinus and a B. indica.

Kandy District
All ten sites in Kandy, except one had infected ectoparasites. Five sites had infected small mammals and ectoparasites while four sites had infected ectoparasites though the host was free of the bacteria ( Table 1).
Of the 118 small mammals, 41 individuals of R. rattus, B. indica, M. fernandoni and Golunda ellioti were infested with external parasites while B. bengalensis and S. murinus were not. Of the infested 41 small mammals, 20 (48.8%) had ectoparasites positive for Rickettsia spp., out of which, 3 were positive for host blood as well. Two Rickettsia positive small mammals had Rickettsia negative ectoparasites. Both hosts and ectoparasites were negative in the other 19 ectoparasite infested small mammals. Of the 77 non-infested small mammals 4 were positive for Rickettsia and 73 were negative. There were 37 small mammals without blood samples, of which 11 had parasites. Eight of them were positive for Rickettsia.
Of the ectoparasites, R. haemaphysaloides, H. spinigera and I. ceylonensis ticks and X. cheopes and S. aporus fleas were carrying Rickettsia. The number of gltA copies in them ranged from 50 to 1.9x10 7 . Rhipicephalus haemaphysaloides nymph individuals, a pool of two nymphs and pools of R. haemaphysaloides larvae, Nymphs of H. spinigera and X. cheopes fleas were carrying large quantities of Rickettsia. Three small mammals were positive for both host blood and parasites. Rickettsia positive ectoparasites were collected from R. rattus, B. indica, M. fernandoni and G. ellioti. Parasites with highest quantity of Rickettsia were from R. rattus and B. indica (Appendix 2).

DISCUSSION
This is the first extensive study to report small mammal species and their ectoparasites involved in transmission of Rickettsia in Sri Lanka. Only few small-scale studies have been carried out on reservoir and vector species of Rickettsia in Sri Lanka previously. During 1930s and 1940s prevalence of murine typhus among rats and the role of thromboculid mites in transmission of scrub typhus have been studied by Sri Lankan researches (Premaratna, 2011). Two latest studies reported seroprevalence of Rickettsia among two dog populations (Nanayakkara et al., 2013) and presence of SFG Rickettsia in Amblyomma ticks collected from Wild mammals (Liyanaarachchi et al., 2012).
Prevalence of Rickettsia among small mammals were similar in both districts studied here, but only R. rattus were carrying Rickettsia in Kurunegala while three species, R. rattus, B. indica and M. fernandoni were Rickettsia positive in Kandy. This difference however, could be accounted for the high small mammal and ectoparasite diversity in selected sites in Kandy. Sites in Peradeniya and Mahakanda were scrublands with minimum human interference. Endemic small mammal Mus fernandoni, Goluna ellioti and I. ceylonensis, D. auratus ticks and S. aporus fleas were found in these sites. Outside Sri Lanka, Rickettsia antibodies have been detected among R. rattus and M. musculus from Philipines (Camer et al., 2012), R. rattus from Brazil (Milagres et al., 2013) and B. indica from Thailand (Okabayashi et al., 1996). Okabayashi reported high prevalence of antibodies among B. indica suggesting it to be a reservoir host for SFG Rickettsia. Studies from outside Sri Lanka also support the importance of R. rattus and B. indica as carriers of Rickettsia (Okabayashi, 1996;Camer et al., 2000;Coleman et al., 2003;Kim et al, 2006), however, this is the first report of endemic Mus fernandoni carrying Rickettsia.
Rhipicephalus haemaphysaloides and X. cheopes were the most abundant tick and flea species found from both districts. Both species were positive for Rickettsia in both districts. Haemaphysalis spinigera, I. ceylonensis ticks and S. aporus fleas were also positive in Kandy district. In other regions of the world Rickettsia has been detected from species of Aponomma, Amblyomma, Dermacentor, Hyalomma and Rhipicephalus (Merhej and Raoult, 2011). Sixty three percent of Aponomma hydrosouri, considered as a reservoir for R. honei, were reported as positive for Rickettsia in a study done in Australia (Stenos et al., 2003). In northern Germany 33.3% of I. ricinus, a reservoir for R. Helvetica was reported positive (Schicht et al., 2012), and 41.2% of Amblyomma americanum was carrying Rickettsia in USA (Mixson et al., 2006). Of flea species, X. cheopes is considered a reservoir and the main vector for R. typhi, which has been isolated from X. cheopes in several studies (Laudisoit et al., 2014).
When considering the Rickettsia quantity, R. haemaphysaloides, H. spinigera ticks and X. cheopes fleas are important as they carry large number of Rickettsia. Maximum quantities carried by them were 10 7 , 10 5 , and 10 5 , respectively. Reports of Rickettsia quantity in reservoir animals and vectors are scarce, but similar quantities have been reported in other studies as well. One study has reported 10 6 to 10 7 copies of R. rickettsia in Amblyomma ticks (Eremeeva et al., 2003). The first report of Rickettsia quantity from tick vectors was from northern Germany with a 33.3% prevalence of Ixodes ricinus, a reservoir for R. helvetica had maximum quantity of Rickettsia in larvae, nymphs and adults were 5x10 4 , 8.5x10 6 and 2x10 7 , respectively (Schicht et al., 2012). Present study, reports maximum of 10 7 Rickettsia from R. haemaphysaloides Nymphs and 10 5 from larvae of the same species..
Of the 18 sites sampled, only three sites were free of infected small mammals or ecoparasites. This shows that the distribution of the pathogen is very high. Though rickettsiosis is not a life threatening disease in most cases, it can lead to severe or fatal disease in vertebrate animals as well as in humans (Azad and Beard, 1998;Costa et al., 2002;Nadchatram, 2008). The distribution of this pathogen in the reservoirs is useful in control of the disease and for taking precautionary measures in high rick localities.