LETTERS

Acknowledgments viron Microbiol. 2004;70:3733–5. DOI: scrub (3). Two amplifi cation We thank the following persons and 10.1128/AEM.70.6.3733-3735.2004 reactions were performed, a real-time institutions for their assistance in obtain- quantitative PCR with a probe target- Address for correspondence: Julie R. Sinclair, ing and identifying specimens: Amy Han- ing the O. tsutsugamushi 47-kDa outer CDC Quarantine Station–Philadelphia, P.O. cock-Ronemus, Paul Mead, Ted Nuttall, membrane protein gene with appropri- Box 144, Essington, PA 19029, USA; email: Brigette Husband, and Kerry Pollard; the ate primers and probes (4) and a stan- [email protected] Department of Parasitology, College of dard PCR targeting a 372-nt fragment Veterinary Medicine, University of Penn- of the 56-kDa protein gene (3). sylvania; the Schuylkill Wildlife Rehabili- Buffy coat samples from 11 tation Center; and the Philadelphia Depart- (17.5%) patients were positive for O. ment of Public Health. tsutsugamushi in the real-time quanti- tative PCR and 56-kDa gene Julie R. Sinclair, Alisa Newton, PCR (Table). All 11 patients were Keith Hinshaw, George Fraser, Genotyping from Vientiane or Vientiane Prov- Patrina Ross, Esther Chernak, of ince. PCR products for the 56-kDa Caroline Johnson, tsutsugamushi from gene fragments were purifi ed and se- quenced as described (3). Comparison and Nancy Warren Humans with Scrub Author affi liations: Centers for Disease (3,5) of amplicons for the 11 patients Control and Prevention, Atlanta, Geor- Typhus, Laos with each other and with GenBank gia, USA (J.R. Sinclair); Philadelphia Zoo, sequences identifi ed 6 genotypes. Per- To the Editor: Rickettsial dis- Philadelphia, Pennsylvania, USA (A. New- centages of nucleotide sequence simi- eases have been only recently identi- ton, K. Hinshaw); Pennsylvania Bureau of larity with other sequences available in fi ed as underrecognized but important Laboratories, Lionville, Pennsylvania, USA GenBank ranged from 95.9% to 100% causes of of unknown origin (G. Fraser, N. Warren); and Philadelphia (Table). Interpretation of our results in Laos. In 2006, 63 (14.8%) of 427 Department of Public Health, Philadelphia was also supported by recent phyloge- adults with negative blood cultures (P. Ross, E. Chernak, C. Johnson) netic studies that compared sequences admitted to Mahosot Hospital in Vien- of the entire 56-kDa type-specifi c anti- DOI: 10.3201/eid1409.071690 tiane had , an infection gen gene of isolates from Thailand (6). caused by Orientia tsutsugamushi and LaoUF238 and LaoUF220 genotypes References transmitted by the bite of larval trom- clustered with those of strains related biculid (1). O. tsutsugamushi to the Karp serotype, and LaoUF136 1. Farlow J, Wagner DM, Dukerich M, Stan- is characterized by a wide antigenic ley M, Chu M, Kubota K, et al. Francisella and LaoUF187 clustered with geno- tularensis in the United States. Emerg In- diversity, and isolates are convention- types of strains related to the Gilliam fect Dis. 2005;11:1835–41. ally classifi ed on the basis of reactiv- serotype (2). Other genotypes found in 2. Feldman KA. . J Am Vet Med ity with hyperimmune serum against this study were grouped in 2 clusters Assoc. 2003;222:725–30. DOI: 10.2460/ prototype strains (e.g., Karp, Kato, javma.2003.222.725 that contained genotypes identifi ed in 3. Dennis DT, Inglesby TV, Henderson HA, Gilliam, Kawasaki, Kuroki, or Shimo- Thailand (5) and Taiwan (7) that have Bartlett JG, Ascher MS, Eitzen E, et al. goshi). The 4 hypervariable regions not been linked to a reference serotype Tularemia as a biological weapon. JAMA. within the 56-kDa type-specifi c anti- (Table). 2001;285:2763–73. DOI: 10.1001/jama. gen of O. tsutsugamushi, which is lo- 285.21.2763 Detection of O. tsutsugamushi in 4. Anda P, Segura del Pozo J, Diaz Garcia cated on the outer membrane surface, humans in Laos provides useful infor- JM, Escudero R, Garcia Pena FJ, Lopez are considered to play an essential role mation on genotypes prevalent in this Velasco MC, et al. Waterborne outbreak of in type strain assignment (2). country. Our results were confi rmed tularemia associated with crayfi sh fi shing. In the Lao study (1), in addition Emerg Infect Dis. 2001;7:575–82. by using 2 target genes in 2 PCRs. No 5. Dvorak P. Health offi cials vigilant for ill- to acute-phase serum samples, a 5-mL differences were found between the ness after sensors detect on mall. blood sample anticoagulated with number of days of fever in 11 PCR- Washington Post. 2005 Oct 2 [cited 2008 EDTA was collected at admission from positive patients and number of days Mar 26]. Available from http://www. all patients. After centrifugation, buffy washingtonpost.com/wp-dyn/content/ of fever in 52 PCR-negative patients. article/2005/10/01/AR2005100101209. coat of the serum sample was removed However, the PCR-negative patients html and stored at –80°C (1). DNA was ex- may not have had bacteremia at the 6. Petersen JM, Schriefer ME, Gage KL, tracted from buffy coat samples of 63 time of sample collection. Montenieri JA, Carter LG, Stanley M, et patients whose conditions were diag- al. Methods for enhanced culture recov- Diversity of O. tsutsugamushi ery of . Appl En- nosed by imunofl uorescence assay as genotypes found in Laos includes

Emerging Infectious Diseases • www.cdc.gov/eid • Vol. 14, No. 9, September 2008 1483 LETTERS genotypes closely related to genotypes gene in isolates from Laos might limit 56-kDa protein–encoding gene and from Thailand and Taiwan. This diver- sensitivity and specifi city of serologic of other genes of O. tsutsugamushi sity raises doubt about usual concepts methods. A recent study showed that would help to better characterize the because it has been thought that O. addition of a serotype to the panel of new genotypes identifi ed in our study tsutsugamushi genotypes are restrict- O. tsutsugamushi used for and their relationship with known se- ed to specifi c geographic areas and to testing improved sensitivity of anti- rotypes. Expanding the panel of anti- specifi c vectors (8). Furthermore, body detection in patients in Thailand gens used to test patients suspected of these results might have clinical reper- (10). We demonstrated that, in analy- having scrub typhus to take into ac- cussions because sequence variations sis of sera in the diagnosis of scrub count local antigenic diversity would within the 56-kDa protein gene corre- typhus contracted in Laos, antigen improve sensitivity of serologic as- late with antigenic diversity of geno- pools should contain at least Karp and says for this disease. types of O. tsutsugamushi. This fi nd- Gilliam strain antigens. Furthermore, ing is supported by data for sequences new genotypes identifi ed in patients Acknowledgments of the entire 56-kDa gene of different in Laos might be related to previously We thank Khalid El Karkouri for help isolates (6) and for monoclonal and unrecognized type strains. However, with phylogenetic studies; the patients, human and animal polyclonal antibod- cross-reactivity with Gilliam, Kato, Vimone Soukkhaseum, Khamphong Phi- ies used to map antigenic differences and Kawasaki serotypes enabled se- asakha, Surn Soukkhaseum, Khamthavi among isolates with known sequence rologic diagnosis in the initial study, Frichithavong, Vang Chu, Valy Keol- variations (9). including 1 patient infected with a ouangkhot, Bertrand Martinez-Aussel, Although our data are prelimi- Karp-related bacteria (1). Ko Chang, Chirapha Darasavath, Oud- nary, diversity of nucleotide sequenc- Phylogenetic studies based on ayvone Rattanavong, Siho Sisouphone, es of the 56-kDa protein–encoding larger fragments of sequences of the Mayfong Mayxay, Sisouphane Vidamaly,

Table. Serologic results for 11 patients from Laos positive by real-time quantitative PCR for the 47-kDa outer membrane protein and a PCR for a fragment of the 56-kDa protein–encoding gene of Orientia tsutsugamushi* GenBank accession no. Highest % identity with IgG/IgM titers, acute phase–late phase of 56-kDa–amplified GenBank sequences Patient Gilliam Kato Kawasaki gene fragment (reference) LaoUF74 64/1,048– 128/256–128/256 128/256–128/256 EU168798 These 4 genotypes showed 64/1,048 100% identity with each other LaoUF317 2,048/1,024–ND 128/1,024–ND 2,048/1,024–ND EU168799 and 96.6% with strain BB23 LaoUF366 2,048/512– 128/32–128/32 2,048/512– EU168797 from Thailand (5) 2,048/512 2,048/512 LaoUF396 256/1,024– 128/64–128/64 256/1,024– EU168796 256/1,024 256/1,024 LaoUF220 2,048/512– 2,048/512– 1,024/256– EU168801 These 2 genotypes showed 2,048/1,024 2,048/512 2,048/512 100% identity with UT150 and LaoUF244 1,024/128–ND 256/64–ND 128/64–ND EU168800 UT332, which are related to Karp serotype isolates (6), BB29 from patients in Thailand (5), and TWyU81 from chiggers in Taiwan (7) LaoUF83 128/1,024–ND 0/1,024–ND 0/128–ND EU168795 95.9% with BB23 from Thailand (5) LaoUF136 2,048/1,024– 2,048/1,024– 2,048/1,024– EU168804 99.5% with UT144, UT125, and 2,048/1,024 2,048/1,024 2,048/1,024 UT196 from Thailand, which are related to Gilliam serotype isolates (6) LaoUF186 128/1,024– 128/1,024– 128/1,024– EU168805 100% with BB23 from 2,048/1,024 2,048/1,024 2,048/1,024 Thailand (5) LaoUF187 256/1,024– 256/0–256/0 0/0–256/1,024 EU168803 100% with UT144, UT125, and 256/1,024 UT196 from Thailand (6) LaoUF238 2,048/1,024– 2,048/1,024– 0/512–256/512 EU168802 100% with UT219, UT395, 2,048/1,024 2,048/1,024 UT221, and UT213, which are related to Karp serotype–related isolates (7), and FPW2031 from Thailand (6) *Ig, immunoglobulin; ND, not done. Specific microimunofluorescence assay (IFA) was performed in Marseille, France, by using whole-cell antigens of O. tsutsugamushi serotypes Kato, Gilliam, and Kawasaki. Serotype Karp was not available for serologic assays. An IFA result was positive if 1) titers were >128 for IgG and >64 for IgM, 2) seroconversion was observed in a convalescent-phase serum sample when an acute-phase serum sample was negative, or 3) there was a >4-fold increase in titers between acute- and convalescent-phase serum samples (0 indicates a titer <25).

1484 Emerging Infectious Diseases • www.cdc.gov/eid • Vol. 14, No. 9, September 2008 LETTERS

Mayboun Heuangvongsy, Chanpheng 4. Jiang J, Chan TC, Temenak JJ, Dasch GA, Clindamycin- Thammavong, Bouachanh Rasachack, Ching WM, Richards AL. Development of a quantitative real-time polymerase Resistant Clone of Bounkong Syhavong, Nicholas J. White, chain reaction assay specifi c for Orien- Suriyasack Thongpaseuth, Anisone tia tsutsugamushi. Am J Trop Med Hyg. Clostridium diffi cile Changthongthip, Viengmone Davong, 2004;70:351–6. PCR Ribotype Olay Lattana, Manivanh Vongsouvath, 5. Fournier PE, Siritantikorn S, Rolain JM, Suputtamongkol Y, Hoontrakul S, 027, Europe Kai-amporn Keopaseuth, Sengmani Charoenwat S, et al. Detection of new Symanivong, Viengmala Sihalath, and genotypes of Orientia tsutsugamushi in- To the Editor: Since 2003, out- Alatsany Chandara for participating in fecting humans in Thailand. Clin Micro- breaks of Clostridium diffi cile–as- the study; and Ponmek Dalaloy and Som- biol Infect. 2008;14:168–73. 6. Blacksell SD, Luksameetanasan R, Ka- sociated disease (CDAD) associated mone Phounsavath for support. lambaheti T, Aukkanit N, Paris DH, with the emergence of a hypervirulent This study was supported by the McGready R, et al. Genetic typing of strain have been reported worldwide the 56-kDa type-specifi c antigen gene of Wellcome Trust–Mahosot Hospital–Ox- contemporary Orientia tsutsugamushi iso- (1,2; www.eurosurveillance.org/em/ ford Tropical Medicine Research Collabo- lates causing human scrub typhus at two v12n06/1206-221.asp). This strain has ration, which was supported by the Well- sites in north-eastern and western Thai- been associated with increased dis- come Trust of Great Britain. land. FEMS Immunol Med Microbiol. ease severity and attributable mortal- 2008;52:335–42. 7. Qiang Y, Tamura A, Urakami H, Makisaka ity. Patients infected with C. diffi cile 027 fail to respond to metronidazole Philippe Parola, Y, Koyama S, Fukuhara M, et al. Phylo- genetic characterization of Orientia tsut- therapy (1). Several typing methods Stuart D. Blacksell, sugamushi isolated in Taiwan according to have been applied to further charac- Rattanaphone Phetsouvanh, the sequence homologies of 56-kDa type- terize C. diffi cile PCR ribotype-027, Simaly Phongmany, specifi c antigen genes. Microbiol Immu- including pulsed-fi eld gel electropho- Jean-Marc Rolain, nol. 2003;47:577–83. 8. Kawamura A, Tanaka H. Tsutsugamushi resis (PFGE) (North American pulsed Nicholas P.J. Day, disease: an overview. Tokyo: University fi eld type 1) and restriction enzyme Paul N. Newton, of Tokyo Press; 1995. analysis (REA) (BI). PFGE and REA and Didier Raoult 9. Seong SY, Kim MK, Lee SM, Odgerel Z, are widely used in the United States; Author affi liations: World Health Organiza- Choi MS, Kim IS, et al. Neutralization epitopes on the antigenic domain II of the PCR ribotyping is more commonly tion Collaborative Center for Rickettsial Dis- Orientia tsutsugamushi 56-kDa protein used throughout Europe. More re- eases and Other Arthropod Borne Bacte- revealed by monoclonal antibodies. Vac- cently, 2 multiple-locus variable-num- rial Diseases, Marseille, France (P. Parola, cine. 2000;19:2–9. DOI: 10.1016/S0264- ber tandem-repeat analysis (MLVA) J.-M. Rolain, D. Raoult); Mahosot Hospital, 410X(00)00167-5 10. Suttinont C, Losuwanaluk K, Niwatayakul protocols have been applied to type Vientiane, Laos (S.D. Blacksell, R. Phetsou- K, Hoontrakul S, Intaranongpai W, Silpa- C. diffi cile, and these proved more vanh, S. Phongmany, N.P.J. Day, P.N. New- sakorn S, et al. Causes of acute, undifferen- discriminatory compared to other ton); University of Oxford, Oxford, United tiated, febrile illness in rural Thailand: re- methods (3,4). Furthermore, MLVA Kingdom (S.D. Blacksell, N.P.J. Day, P.N. sults of a prospective observational study. Ann Trop Med Parasitol. 2006;100:363– can subgroup geographically diverse Newton); and Mahidol University, Bangkok, 70. DOI: 10.1179/136485906X112158 027 isolates (G. Killgore et al., unpub Thailand (S.D. Blacksell, N.P.J. Day) data) as well as 027 isolates that are Address for correspondence: Didier Raoult, DOI: 10.3201/eid1409.071259 common to 1 institution (5). Unité des Rickettsies, Centre National de la References We reported a case of C. diffi cile Recherche Scientifi que–Institut de Recherche PCR 027 in Ireland, where the isolate 1. Phongmany S, Rolain JM, Phetsouvanh pour le Développement, Unité Mixte de had an identical antibiogram profi le R, Blacksell SD, Soukkhaseum V, Rasa- Recherche 6236, World Health Organization compared with those strains reported chack B, et al. Rickettsial infections and Collaborative Center for Rickettsioses and Other across Europe (6,7) (i.e., resistant to fever, Vientiane, Laos. Emerg Infect Dis. Arthropod Borne Bacterial Diseases, Faculté de 2006;12:256–62. fl uoroquinolones and erythromycin, 2. Tamura A, Yamamoto N, Koyama S, Médecine, 27 Bd Jean Moulin, 13005 Marseille, susceptible to clindamycin). We have Makisaka Y, Takahashi M, Urabe K, et France; email: [email protected] subsequently identifi ed C. diffi cile al. Epidemiological survey of Orientia tsutsugamushi distribution in fi eld rodents 027 in 6 more healthcare settings. To in Saitama Prefecture, Japan, and discov- The opinions expressed by authors con- date >100 Irish C. diffi cile 027 isolates ery of a new type. Microbiol Immunol. tributing to this journal do not necessar- have been characterized by analysis of ily refl ect the opinions of the Centers for 2001;45:439–46. their antibiogram profi les, toxinotyp- 3. Mahajan SK, Rolain JM, Kashyap R, Bak- Disease Control and Prevention or the shi D, Sharma V, Prasher BS, et al. Scrub institutions with which the authors are af- ing, and 16S–23S rDNA PCR ribotyp- typhus in Himalayas. Emerg Infect Dis. fi liated. ing. All C. diffi cile 027 isolates were 2006;12:1590–2. resistant to moxifl oxaxin, gatifl oxacin,

Emerging Infectious Diseases • www.cdc.gov/eid • Vol. 14, No. 9, September 2008 1485