Rickettsia sp. in Ixodes granulatus Ticks, Japan

To the Editor: The genus Rickettsia consists of obligate intracellular bacteria that cause spotted fever and typhus fever; these bacteria are usually transmitted by an arthropod vector. We report isolation of a Rickettsia honei–like organism from the Ixodes granulatus tick; this organism may be a causative agent of rickettsiosis in Japan. Serotyping and DNA-sequencing analysis distinguished this I. granulatus isolate from previously reported Rickettsia spp.


Rickettsia sp. in Ixodes granulatus
Ticks, Japan To the Editor: The genus Rickettsia consists of obligate intracellular bacteria that cause spotted fever and typhus fever; these bacteria are usually transmitted by an arthropod vector. We report isolation of a Rickettsia honei-like organism from the Ixodes granulatus tick; this organism may be a causative agent of rickettsiosis in Japan. Serotyping and DNA-sequencing analysis distinguished this I. granulatus isolate from previously reported Rickettsia spp. During 2004During -2005, an investigation of rickettsiosis was conducted in Okinawa Prefecture in the southernmost part of Japan, an area known to be inhabited by I. granulatus, a parasitic tick commonly found on small mammals. A total of 43 I. granulatus ticks (3 larvae, 27 nymphs, 8 adult females, and 5 adult males) were collected from small mammals (Rattus rattus, R. norvegicus, Suncus murinus, Mus calori, and Crocidura watasei) for the present study. For the isolation of Rickettsia spp., the cell line L929 was used as previously described (1). A total of 13 isolates, designated as strains GRA-1 to GRA-13, were obtained from 11 ticks (1 fed larva, 5 fed nymph, 1 fed adult female, 1 fed adult male, 1 unfed nymph molted from engorged larva, 2 unfed adult females molted from engorged nymphs) and from 1 pool of eggs and 1 larva derived from the engorged female tick. Serotyping was performed by using a microimmunoperoxidase approach according to the method described by Philip et al. (2); we used anti-Rickettsia mouse serum and several spotted fever group Rickettsia antigens: 2 of the present isolates (GRA-1 and GRA-2) and 6 known members of the Asian Rickettsia spp. (R. honei, R. japonica, R. asiatica, R. tamurae, R. sibirica, and R. conorii). Differences among antigen reaction titers were calculated, and the results are given as the specifi city difference (SPD) value. The SPD value between the present isolates and R. honei was 0 or 1, whereas the SPD values were >3 for the other spotted fever group Rickettsia spp. (Table). According to the criteria for serotyping (2), we assumed the isolates to be of the same serotype when the SPD value was <2. In addition to serotyping, a sequencing analysis was performed to genetically characterize the isolates. The archive of DNA sequences has been mostly established for the outer membrane protein A gene (ompA), citrate synthesis gene (gltA), and 17-kDa antigen gene. Thus, we determined these DNA sequences in the isolates and compared the results with those of representative Rickettsia spp. The ompA sequencing analysis showed a DNA sequence of 491 bp in the 6 isolates from I. granulatus (GenBank accession nos. AB444090-AB44095), which yielded the following similarity values: R. slovaca (98.0%), R. honei and Thai tick typhus Rickettsia (97.8%), and R. honei subsp. marmionii (97.6%). Sequencing of the 1,250-bp fragment of gltA of the strain GRA-1 (accession no. AB444098) showed >99% DNA similarity with that of R. sibirica (99.3%), R. slovaca (99.2%), R. conorii (99.2%), R. honei (99.1%), and certain types of Rickettsia spp. Moreover,17-kDa antigen gene sequencing analysis of a 392-bp fragment of the strain GRA-1 (accession no. AB444097) showed the highest levels of sequencing similar-ity value with R. honei and Thai tick typhus Rickettsia (99.5%) compared with those of the sequences of other deposited Rickettsia spp. Comprehensive analyses led us to presume that the isolate GRA-1 from I. granulatus was a genetic variant of R. honei, although further studies are necessary to better defi ne the taxonomic position of our isolates.
The vector for R. honei was presumed to be ixodid ticks: I. granulatus in Thailand; Amblyomma cajennense in Texas, USA; and Aponomma hydrosauri in Australia (3)(4)(5). Lane et al. reported that a Rickettsia organism from a Haemaphysalis tick was closely related to R. honei in Australia (6). In the present study, we observed that the Rickettsia organism was maintained in the tick after molting. Moreover, Rickettsia organisms were also isolated from egg and unfed larva. These preliminary fi ndings may suggest that I. granulatus is a possible vector for the R. honei-like bacterium in Japan.
Recently, a Rickettsia sp. was found in I. granulatus ticks; its proposed designation was unclassifi ed Rickettsia IG-1, according to DNA sequencing from specimens obtained in Taiwan (7). With respect to the DNA sequences of gltA and ompA, our isolates from I. granulatus were identical to the Rickettsia IG-1.
R. honei, a member of the spotted fever group Rickettsia, has been reported as the etiologic agent of Flinders Island spotted fever in Australia (8) and also of Thai tick typhus (3). R. honei is a public health threat for rickettsiosis in these countries. Although the human health implications of the Rickettsia sp. found in this study are not yet known, knowledge from this study will be useful in epidemiologic investigation for rickettsiosis in Japan.

Acknowledgments
We  (Table). Upon capture of the mice, blood samples were taken from the submandibular vein by using Medi-Point animal lancets (Medi-Point International, Inc., Mineola, NY, USA) and stored in sterile micropipette tubes. Samples were stored on ice until shipment to the California Department of Health Services' Viral and Rickettsial Disease Laboratory for processing. P. maniculatus serum samples were examined for immunoglobulin (Ig) G antibodies to the SNV nucleocapsid protein by ELISA with Centers for Disease Control and Prevention reagents (5).
Detailed information regarding SNV prevalence, sampling location, and sampling effort is presented in the Table. We compare our 2007 data with data collected in 1994 by Jay et al. (2) because 1994 was the only other year when all 4 islands used in our study were sampled. Graham and Chomel (4) also collected data from San Miguel Island and Santa Rosa Island in 1995 and 1996 (the use of the average prevalence from 1995 and 1996 for these 2 islands does not change any of our results).
There was no signifi cant difference in prevalence of SNV antibodies between our 2007 results and the prevalence found by Jay et al. (2) in 1993-1994 (paired t test t = 0.13, 3; df = 3; p = 0.91). Overall, 36 male and 42 female mice were captured; the sex of captured animals was independent of SNV infection (9 males and 6 females positive for SNV; test of independence χ 2 = 0.28, 1 df, p = 0.59). We captured only 2 subadult mice on islands where we also detected antibodies to SNV; 1 mouse tested positive, the other tested negative. Although our sample sizes precluded detecting very low rates of SNV infection with confi dence on Santa Barbara and East Anacapa Islands, the consistency of our results with those of Jay et al. (2) suggests that our sampling was suffi cient for comparative purposes.
Several studies now indicate the importance of long-term surveillance of SNV prevalence in wild rodent populations for understanding the factors that may contribute to outbreaks of human disease, e.g. (6). These studies often document the generally positive, though often temporally delayed, relationship between population density of P. maniculatus and seroprevalence for SNV (7). Our results suggest a high degree of temporal stability in prevalence of antibodies to SNV in P. maniculatus on the Channel Islands, despite considerable variation in host population density between earlier studies and ours (4,8). Although we cannot know the prevalence of SNV among P. maniculatus on the Channel Islands in periods between the studies by Jay et al. (2), Graham and Chomel (4), and our own, SNV prevalence on these islands is quite similar to levels previously recorded both for islands with relatively low prevalence  (2) and are included for comparison puposes. †East Anacapa: 34º00'56"N/119º21'49"W. ‡San Miguel: 34º02'18"N/120º20'54"W. §Santa Barbara: 33º28'30"N/119º02'12"W. ¶Santa Rosa: 34º00'03"N/120º03'30"W.