Identification of rickettsiae from wild rats and cat fleas in Malaysia

Medical and Veterinary Entomology (2014) 28 (Suppl. 1), 104 108 SHORT COMMUNICATION Identification of rickettsiae from wild rats and cat fleas in Malaysia S. T. T A Y 1, A. S. MOKHTAR 1, K. C. L OW 2, S. N. MOHD ZAIN 3, J. JEFFERY 4, N. ABDUL AZIZ 4 and K. L. KHO 1 1 Department of Medical Microbiology, Tropical Infectious Diseases Research and Education Centre, Faculty of Medicine, University of Malaya, Kuala Lumpur, Malaysia, 2 Laboratory Animal Resource Unit, Faculty of Medicine, Universiti Kebangsaan Malaysia, Kuala Lumpur, Malaysia, 3 Institute of Biological Sciences, Faculty of Science, University of Malaya, Kuala Lumpur, Malaysia and 4 Department of Parasitology, Faculty of Medicine, University of Malaya, Kuala Lumpur, Malaysia Abstract. Rickettsioses are emerging zoonotic diseases reported worldwide. In spite of the serological evidence of spotted fever group rickettsioses in febrile patients in Malaysia, limited studies have been conducted to identify the animal reservoirs and vectors of rickettsioses. This study investigated the presence of rickettsiae in the tissue homogenates of 95 wild rats and 589 animal ectoparasites. Using PCR assays targeting the citrate synthase gene (glta), rickettsial DNA was detected in the tissue homogenates of 13 (13.7%) wild rats. Sequence analysis of the glta amplicons showed 98.6 100% similarity with those of Rickettsia honei/r. conorii/r. raoultii (Rickettsiales: Rickettsiaceae). Sequence analysis of outer membrane protein A gene (ompa) identified Rickettsia sp. TCM1 strain from two rats. No rickettsia was detected from Laelaps mites, Rhipicephalus sanguineus and Haemaphysalis bispinosa ticks, and Felicola subrostratus lice in this study. R. felis was identified from 32.2% of 177 Ctenocephalides felis fleas. Sequence analysis of the glta amplicons revealed two genotypes of R. felis (Rf31 and RF2125) in the fleas. As wild rats and cat fleas play an important role in the enzoonotic maintenance of rickettsiae, control of rodent and flea populations may be able to reduce transmission of rickettsioses in the local setting. Key words. Cat fleas, reservoirs, rickettsioses, wild rats, Malaysia. Rickettsiae are Gram-negative obligate intracellular bacteria which have been associated with a variety of infections including Rocky Mountain spotted fever, epidemic typhus, murine typhus, Mediterranean spotted fever etc. Arthropods such as ticks, fleas, mites and lice are important vectors for transmission of rickettsioses. As the organisms are difficult to culture directly from samples, molecular tools such as PCR and sequence analysis have been used to facilitate the identification and characterization of rickettsial species. To date, 26 rickettsial species with validated and published names have been reported; Rickettsia sibirica, R. heilongjiangensis, R. japonica, R. conorii, R. honei, R. aeschlimannii, R. raoultii, R. slovaca, R. helvetica, R. massiliae and R. monacensis are amongst those tick-borne rickettsiae that have been reported in the Asia region (Parola et al., 2013). Rickettsia felis, the causative agent of flea-borne spotted fever, has been recently recognized as an emerging global health threat (Pérez-Osorio et al., 2008). First described in the Ctenocephalides felis cat flea, R. felis has a wide cosmopolitan distribution, detected from arthropods in > 20 countries and five continents (Parola, 2011). As the clinical manifestations of R. felis infections are non-specific and difficult to differentiate from other causes of febrile disease such as murine typhus and dengue (Pérez-Osorio et al., 2008), disease caused by R. felis infection is potentially underdiagnosed. The importance of R. felis as a common cause of febrile diseases in sub-saharan rural Africa has been recently documented (Mediannikov et al., 2013). Current data on the prevalence and type of rickettsioses in Malaysia are scarce. Although evidence is available of natural infection of spotted fever group rickettsiae in the wild Correspondence: Sun T. Tay, Department of Medical Microbiology, Faculty of Medicine, University of Malaya, 50603 Kuala Lumpur, Malaysia. Tel.: 03-79676676; Fax: 03-79676660; E-mail: tayst@um.edu.my 104 2014 The Royal Entomological Society

Rickettsiae from wild rats and cat fleas 105 rodents (Rattus and Tupaia species) and Haemaphysalis ticks in Malaysia (Tay et al., 1996, 1998), attempts to isolate rickettsiae have not been successful. A high antibody prevalence to R. honei (TT118 strain) spotted fever group rickettsia has been reported in febrile patients in rural and urban areas in Malaysia (Tay et al., 2000, 2003); however, information is lacking concerning the animal reservoirs and vectors of this rickettsial species. Rickettsia honei has also been reported in patients from Thailand (Jiang et al., 2005), a neighbouring country of Malaysia. Other than R. honei, a rickettsial species closely related to R. raoultii has been reported from an Amblyomma tick from Thailand (Doornbos et al., 2013). Using specific polymerase chain reaction (PCR) assays, R. felis DNA has been detected in fleas in our previous study (Mokhtar & Tay, 2011). To identify risk, and emergence of new rickettsial organisms, mapping of animal hosts and arthropod vectors in a particular area is essential. Rats are competent reservoir hosts of several zoonotic pathogens and have been used as a sentinel organism for rickettsial studies (Psaroulaki et al., 2010; Kuo et al., 2012). Furthermore rats can bring diseases to humans by increasing exposure to potentially infected ticks, mites and fleas, and together, they play important roles as amplifying hosts and transporters of rickettsiae. Hence, this study was performed with the aim to assess wild rats and animal ectoparasites as potential reservoirs for rickettsioses in Malaysia. A total of 95 wild rats (58 Rattus diardii and 37 R. norvegicus) caught at the wet markets in Kuala Lumpur and Pulau Pinang from January 2008 to December 2011 were included in this investigation. Rodents were caught using live traps, and anaesthetized using an ether-charged chamber. Rats were dissected and organs (kidney, liver, spleen and heart) were harvested aseptically and kept at 80 C prior to processing. The organs were homogenized using a pestle and mortar in phosphate-buffered saline. DNA was extracted from 20 mg of animal tissues using a QIAamp DNA Mini Kit (Qiagen, Hilden, Germany). Ectoparasites were collected from the rats and pooled for screening of rickettsiae. In addition, ticks and lice collected from cats and dogs from three sampling sites in Malaysia (Kuala Nerang, Pendang and Ampang) during January to August 2010 were investigated in this study (Table 1). A total of 177 fleas collected from several locations in Kuala Lumpur and Selangor (Hulu Langat and Gombak), Malaysia were also included in this study. The ectoparasites were kept in micro-centrifuge tubes containing 70% alcohol for identification by entomologists. A protocol described by Alekseev et al. (2001) was used to prepare a DNA template from the ectoparasites. Briefly, each ectoparasite was immersed in 100 μl of 0.7 m ammonium hydroxide and boiled for 20 min. The extracted DNA was then resuspended in 10 μl of sterile distilled water prior to amplification. A PCR assay utilizing primer pair CS-78 and CS-323 was used to amplify the rickettsial citrate synthase (glta) gene fragment (Labruna et al., 2004). The positive samples were further subjected to amplification using (a) primers CS-239 and CS-1069, targeting a 830-bp fragment of the citrate synthase (glta-1) gene fragment (Labruna et al., 2004); (b) primers Rr190_70p and Rr_602n, targeting a 532-bp fragment of the rickettsial 190-kDa outer membrane protein gene (ompa) (Regnery et al., 1991) and (c) primers 120-M59f and 120-807, targeting an approximately 860-bp fragment of the rickettsial 135-kDa outer Table 1. Molecular detection of rickettsiae from rats and animal ectoparasites using glta PCR assay in this study. Sample No. (%) rat positive for rickettsial DNA Wild rats (n = 95) Rattus diardii (n = 58) 9 (15.5) i. PM1 (male, spleen) ii. PM2 (female, spleen, liver) iii. BB7 (female, kidney, spleen) iv. DK2 (male, kidney, spleen, liver) v. KL2 (male, liver, spleen) vi. KL8 (female, spleen) vii. KL18 (female, kidney) viii. PP26 (female, heart) ix. PP34 (female, heart) Rattus norvegicus (n = 37) 4 (10.8) i. DK6 (female, kidney, spleen, liver) ii. PP33 (male, kidney) iii. PP35 (male, heart) iv. PP36 (male, heart) Ectoparasite, animal host (n = 589) Laelaps spp. mite, rats (n = 63) 0 (0) Haemaphysalis bispinosa ticks, cats (n = 7) 0 (0) Felicola subrostratus lice, cats (n = 73) 0 (0) Rhipicephalus sanguineus ticks, dogs 0(0) (n = 165) Heterodoxus spiniger lice, dogs (n = 104) 0 (0) Ctenocephalides felis fleas, cats and dogs 57 (32.2) (n = 177) OmpA was amplified. membrane protein gene (ompb) (Roux & Raoult, 2000). The glta amplicon from R. honei (strain TT118) was cloned into the PCR2.1-TOPO T/A plasmid vector and used as a positive control for PCR assays. For detection of R. felis, DNA extracts from R. felis-positive fleas from a previous study (Mokhtar & Tay, 2011) were used as positive controls. Purification of amplicons was carried out using a LaboPass PCR Purification Kit (Cosmos Genetech, Seoul, Korea) as described by the manufacturer. Amplicons were sequenced using a BigDye terminator cycle sequencing kit (Applied Biosystems, Foster City, CA, U.S.A.) on an ABI-3730 Genetic Analyzer (Applied Biosystems, Foster City, CA, U.S.A.) using their respective PCR primers. Sequences were imported into the BioEdit sequence alignment program and inspected manually (Hall, 1999). The neighbour-joining method of MEGA software (version 5.1) was employed to determine the phylogenetic status of the isolates (Tamura et al., 2011). The reliability of different phylogenetic groupings was evaluated using bootstrap tests (1000 bootstrap replicates). For the phylogenetic study, glta sequences of the type strains were retrieved from the Genbank database. Table 1 shows the detection of rickettsiae in the wild rats investigated in this study. A total of 13 (13.7%) wild rats were positive for rickettsiae by glta PCR assay. Positive specimens were derived from the tissue homogenates of the livers (n = 4), hearts (n = 4), kidneys (n = 5) and spleens (n = 7). Sequence analysis of the glta gene fragments (265 nucleotides) from five rats (PM1, PM2, BB7, DK2 and DK6) was performed using bidirectional sequences. All demonstrated 100% similarity with

106 S. T. Tay et al. 97 55 Rat specimen PP26/PP33 50 43 Rat specimen PP34 Rat specimen KL18 Rat specimen PP36/KL2/KL8 R. honei RB (AF018074) R. conorii NIAID Malish 7 (U59730) 13 R. raoultii strain Elanda23/95 (EU036985) R. raoultii Khabarovsk (DQ365804) Rat specimen PM1/PM2/DK2/DK6/BB7/PP35 (KF963597-KF963601) R. massiliae Mtu 1 (U59719) 7 59 48 40 R. rhipicephali Burgdorfer 3-7-female 6 (U59721) R. parkeri NIAID maculatum 20 (U59732) R. sibirica 246 (U59734) R. africae ESF5 (U59733) 10 9 R. raoultii strain Marne (DQ365803) R. heilongjiangii 054 (AF178034) 41 R. japonica YM (U59724) 47 23 R. slovaca 13B (U59725) 28 60 66 R. montananensis (U74756) R. rickettsii R (Bitterroot) (U59729) R. aeschlimannii MC16 (U59722) R. tamurae strain AT-1 (AF394896) 57 36 R. asiatica strain IO1 (AF394901) R. helvetica C9P9 (U59723) 83 R. akari MK(Kaplan) (U59717) R. australis Phillips (U59718) 99 Flea specimen HL2a (KF963602) Rickettsia sp. RF2125 (AF516333) 94 36 59 R. felis URRWXCal2 (AF210692) Flea specimen HL15c (KF963603) 64 R. felis Rf31 (AF516331) R. prowazekii Breinl (M17149) 99 R. typhi Wilmington (U59714) R. bellii 369L421 (U59716) R. canadensis 2678 (U59713) 0.02 Fig. 1. Unrooted dendrogram showing the phylogenetic position of rickettsiae detected from rat and ectoparasite specimens in this study with other published Rickettsia species (based on comparison of glta sequences). Bootstrap values are indicated at nodes. those of R. honei/r. conorii/r. raoultii (Table 1, GenBank Accession numbers: KF963597-KF963601). Greater variation (ranging from 98.4% to 100% compared with those of R. honei/r. conorii/r. raoultii) was noted for glta sequences derived from the remaining eight rats (based on one-directional sequences). Figure 1 is a dendrogram showing the clustering of the rickettsiae detected from our rat specimens with published R. honei/r. conorii/r. raoultii type strains. Further differentiation of rickettsiae to the species level is not possible owing to the high sequence similarity shared amongst the taxonomically very close related Rickettsia species (as indicated by the low bootstrapping values in the dendrogram). None of the glta-positive rat specimens were positive for amplification using primers targeting glta-1 and ompb. OmpA amplicons were obtained from two kidney homogenates only. The ompa sequences (BB7 and DK2, GenBank Accession numbers: KF963604 and KF963605) were identical to Rickettsia sp. TCM1 strain (GenBank Accession number: AB359459) which is closely related to R. japonica, the causative agent of Japanese spotted fever (Mahara, 2006; Takada et al., 2009)

Rickettsiae from wild rats and cat fleas 107 which has recently been isolated from a H. hystricis tick in Thailand (Takada et al., 2009). At this point, we are not clear as to why the results of glta sequence analysis for the rickettsiae detected from the kidney homogenates of two rats were not identical with those of ompa sequence analysis. Mixed rickettsial infections are suspected in these rats. Additionally, in a multi-template condition as for the tissue homogenates in this study, the difference in the amplification results can be caused by PCR bias whereby certain genes are amplified more efficiently compared with others (Kanagawa, 2003). Probe hybridization may be useful to confirm the presence of two different rickettsiae from the same tissue specimen. We have attempted to infect Vero cells with six rickettsiae-positive rat specimens, however, no rickettsiae were detected from the infected cells after the first passage using glta PCR assays. Table 1 shows the results of the molecular detection of rickettsiae from animal ectoparasites in this study. The ectoparasites collected from the wild rats in this study were mainly spiny rat mites (Laelaps spp.), none of which (processed in 26 pools) were positive for rickettsiae by glta PCR assay. Rickettsial DNA was also not detected from the DNA extracts of Rhipicephalus sanguineus and H. bispinosa ticks, although these ticks have been considered as potential vectors for rickettsial pathogens including R. conorii (Lane et al., 2005; Rovery et al., 2008). So far, no report is available on the importance of H. spiniger and F. subrostratus lice as vectors in the transmission of rickettsioses. None of these lice were positive for rickettsial DNA in this study. Rickettsia felis was detected from 32.2% of 177 C. felis fleas in this study. This detection rate is higher than that of 2.9% reported in a previous study in Malaysia (Mokhtar & Tay, 2011), but similar to the high infection rates in wild cat flea populations in South America (varying from 13.5 to 90%, Labruna et al., 2007) and laboratory cat flea colonies in the United States (varying from 43 to 100%, Reif et al., 2008). Sequence analysis of the glta amplicons from our flea specimens demonstrated two R. felis genotypes. The glta amplicons from 56 of 57 flea specimens (represented by specimen HL2a, GenBank Accession number: KF963602) demonstrated 100% sequence similarity with that of Rickettsia sp. RF2125 (GenBank Accession number: AF516333), which has been reported in fleas in a previous study (Mokhtar & Tay, 2011). The glta amplicon from a single flea specimen (HL15c, GenBank Accession number: KF963603) showed 100% sequence similarity with that of R. felis Rf31 (GenBank Accession number: AF516331) which was first described in fleas from the Thai-Myanmar border (Parola et al., 2003). Amplification of a longer citrate synthase gene (glta-1) sequences of the flea specimens HL2a (GenBank Accession number: KF963606) and HL15c (GenBank Accession number: KF963607) demonstrated 97.7% (708/725 nt) and 97.2% (705/725 nt) similarity when compared with that of R. felis type strain URRWXCal2 (GenBank Accession number: CP000053), respectively. OmpB was only amplified from the flea specimen HL15c (GenBank Accession number: KF963608), and the sequence was 95.0% (756/796 nt) similar to that of R. felis type strain URRWXCal2. Figure 1 shows the phylogenetic placement of the R. felis glta sequences obtained from our flea specimens with published R. felis sequences. As R. felis cross-reacts with typhus and spotted fever group rickettsiae, it may contribute to the high seroprevalence of typhus and spotted fever group rickettsiae observed in our previous serosurveys (Tay et al., 2000, 2003). The absence of R. typhi (causative agent of murine typhus) in this study reflects the absence or very low frequency of R. typhi in fleas from this area. In conclusion, the findings of this study suggest the presence of two different spotted fever group rickettsiae (tentatively identified as R. honei/r. conorii/r. raoultii and Rickettsia sp. TCM1) among wild rats in Malaysia. As the rats were caught from market areas, it is possible for the infection to spill over into the human population through the ectoparasite vectors that feed on both rats and humans. In addition, owing to the frequent feeding behaviour and mobility of fleas, rapid spreading of flea-borne R. felis to human populations is possible. Close association with domestic animals, i.e. dogs and cats and their ectoparasites, can increase human risk of acquiring rickettsial infections. Further research to isolate and characterize the rickettsial organism is important to improve our understanding about human rickettsioses in Malaysia. In addition, control of rodent and flea populations may be able to reduce the transmission of rickettsioses in this part of the world. Acknowledgements This study was funded by grants HIR/E000013-20001 (subprogramme 4), PV086/2011B and RP013/2012A from University of Malaya, Kuala Lumpur, Malaysia. We acknowledge the assistance given by the technical assistance given by S. Samulung, J. Nagappan, N. H. Mohd Isa, N. Shahimin, B. Douadi and K. S. Chai. Author s declaration of interests No competing interests have been declared. References Alekseev, A.N., Dubinina, H.V., Van De Pol, I. & Schouls, L.M. (2001) Identification of Ehrlichia spp. and Borrelia burgdorferi in Ixodes ticks in the Baltic Regions of Russia. Journal of Clinical Microbiology, 39, 2237 2242. Doornbos, K., Sumrandee, C., Ruang-Areerate, T., Baimai, V., Trinachartvanit, W. & Ahantarig, A. (2013) Rickettsia sp. closely related to Rickettsia raoultii (Rickettsiales: Rickettsiaceae) in an Amblyomma helvolum (Acarina: Ixodidae) tick from a Varanus salvator (Squamata: Varanidae) in Thailand. Journal of Medical Entomology, 50, 217 220. Hall, T.A. (1999) BioEdit: a user friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucleic Acids Symposium Series, 41, 95 98. Jiang, J., Sangkasuwan, V., Lerdthusnee, K., Sukwit, S., Chuenchitra, T. & Rozmajzl, P.J. (2005) Human infection with Rickettsia honei, Thailand. Emerging Infectious Diseases, 11, 1473 1475. Kanagawa, T. (2003) Bias and artifacts in multitemplate polymerase chain reactions (PCR). Journal of Bioscience and Bioengineering, 96, 317 323.

108 S. T. Tay et al. Kuo, C.C., Huang, J.L., Lin, T.E. & Wang, H.C. (2012) Detection of Rickettsia spp. and host and habitat associations of fleas (Siphonaptera) in eastern Taiwan. Medical and Veterinary Entomology, 26, 341 350. Labruna, M.B., Whitworth, T., Horta, M.C. et al. (2004) Rickettsia species infecting Amblyomma cooperi ticks from an endemic area for Brazilian spotted fever in the state of Sao Paolo, Brazil. Journal of Clinical Microbiology, 42, 90 98. Labruna, M.B., Ogrzewalska, M., Moraes-Filho, J., Lepe, P., Gallegos, J.L. & López, J. (2007) Rickettsia felis in Chile. Emerging Infectious Diseases, 13, 1794 1795. Lane, A.M., Shaw, M.D., McGraw, E.A. & O Neill, S.L. (2005) Evidence of a spotted fever-like rickettsia and a potential new vector from northeastern Australia. Journal of Medical Entomology, 42, 918 921. Mahara, F. (2006) Rickettsioses in Japan and the Far East. Annals of the New York Academy of Sciences, 1078, 60 73. Mediannikov, O., Socolovschi, C., Edouard, S. et al. (2013) Common epidemiology of Rickettsia felis infection and malaria, Africa. Emerging Infectious Diseases, 19, 1775 1783. Mokhtar, A.S. & Tay, S.T. (2011) Molecular detection of Rickettsia felis, Bartonella henselae,andb. clarridgeiae in fleas from domestic dogs and cats in Malaysia. American Journal of Tropical Medicine and Hygiene, 85, 931 933. Parola, P. (2011) Rickettsia felis: from a rare disease in the USA to a common cause of fever in sub-saharan Africa. Clinical Microbiology and Infection, 17, 996 1000. Parola, P., Sanogo, O.Y., Lerdthusnee, K. et al. (2003) Identification of Rickettsia spp. and Bartonella spp. in fleas from the Thai-Myanmar border. Annals of the New York Academy of Sciences, 990, 173 181. Parola P, Paddock, C.D., Socolovschi, C. et al. (2013) Update on tick-borne rickettsioses around the world: a geographic approach. Clinical Microbiology Reviews, 26, 657 702. Pérez-Osorio, C.E., Zavala-Velázquez, J.E., Arias León, J.J. & Zavala-Castro, J.E. (2008) Rickettsia felis as emergent global threat for humans. Emerging Infectious Diseases, 14, 1019 1023. Psaroulaki, A., Antoniou, M., Toumazos, P. et al. (2010) Rats as indicators of the presence and dispersal of six zoonotic microbial agents in Cyprus, an island ecosystem: a seroepidemiological study. Transactions of the Royal Society of Tropical Medicine and Hygiene, 104, 733 739. Regnery, R.L., Spruill, C.L. & Plikaytis, B.D. (1991) Genotypic identification of rickettsiae and estimation of intraspecies sequence divergence for portions of two rickettsial genes. Journal of Bacteriology, 173, 1576 1589. Reif, K.E., Stout, R.W., Henry, G.C., Foil, L.D. & Macaluso, K.R. (2008) Prevalence and infection load dynamics of Rickettsia felis in actively feeding cat fleas. PLoS One, 3, e2805. Roux, V. & Raoult, D. (2000) Phylogenetic analysis of members of the genus Rickettsia using the gene encoding the outer-membrane protein rompb (ompb). International Journal of Systematic and Evolutionary Microbiology, 50, 1449 1455. Rovery, C., Brouqui, P. & Raoult, D. (2008) Questions on Mediterranean spotted fever a century after its discovery. Emerging Infectious Diseases, 14, 1360 1367. Takada, N., Fujita, H., Kawabata, H., Ando, S., Sakata, A., Takano, A. & Chaithong, U. (2009) Spotted fever group Rickettsia sp. closely related to Rickettsia japonica, Thailand. Emerging Infectious Diseases, 15, 610 611. Tamura, K., Peterson, D., Peterson, N., Stecher, G., Nei, M. & Kumar, S. (2011) MEGA5: molecular evolutionary genetics analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods. Molecular Biology and Evolution, 28, 2731 2739. Tay, S.T., Ho, T.M. & Rohani, M.Y. (1996) Use of hemolymph test for detection of rickettsiae in Malaysian ticks. Southeast Asian Journal of Tropical Medicine and Public Health, 27, 642 644. Tay, S.T., Kaewanee, S., Ho, T.M., Rohani, M.Y. & Devi, S. (1998) Serological evidence of natural infection of wild rodents (Rattus spp and Tupaia glis) with rickettsiae in Malaysia. Southeast Asian Journal of Tropical Medicine and Public Health, 29, 560 562. Tay, S.T., Ho, T.M., Rohani, M.Y. & Devi, S. (2000) Antibodies to Orientia tsutsugamushi, Rickettsia typhi and spotted fever group rickettsiae among febrile patients in rural areas of Malaysia. Transactions of the Royal Society of Tropical Medicine and Hygiene, 94, 280 284. Tay, S.T., Kamalanathan, M. & Rohani, M.Y. (2003) Antibody prevalence of Orientia tsutsugamushi, Rickettsia typhi and TT118 spotted fever group rickettsiae among Malaysian blood donors and febrile patients in the urban areas. Southeast Asian Journal of Tropical Medicine and Public Health, 34, 165 170. Accepted 22 January 2014