Vol. 58, No. 4INFECTION AND IMMUNITY, Apr. 1990, p. 887-8920019-9567/90/040887-06$02.00/0Copyright © 1990, American Society for Microbiology

Causative Agent of Spotted Fever Group Rickettsiosis in JapanTAKANORI OKADA,* YOSHIKI TANGE, AND YUZURU KOBAYASHI

First Department of Internal Medicine, School of Medicine,Ehime University, Shigenobu, Ehime, Japan

Received 26 October 1989/Accepted 13 January 1990

Since 1984, it has been known that spotted fever group rickettsiosis exists in Japan. We isolated three strainsof the causative rickettsiae, designated Katayama, Misaka, and Abe, from patients with the disease and studiedthe characteristics of the isolates. Nude mice and cyclophosphamide-treated mice died after infection with theisolates. However, infected normal mice recovered and acquired immunity. Infected adult male guinea pigs hadfever, a scrotal reaction, and seroconversion. The isolates propagated well in tissue-cultured Vero cells.Analysis by the cross-immunofluorescence antibody method showed that these isolates were closely relatedserologically. To reveal their immunological properties in detail, we produced 21 anti-Katayama monoclonalantibodies. Seven of these antibodies reacted with all representative strains of spotted fever group rickettsiaeused in this study, and five others reacted only with the homologous strain, revealing that the Katayama strainhas a strain-specific antigen(s) different from those of other spotted fever group rickettsiae. Moreover, thesestrain-specific antibodies also reacted with the Misaka and Abe strains. These results demonstrate that thecausative agent of spotted fever group rickettsiosis in Japan is a new serotype of spotted fever group rickettsiae.

Spotted fever group (SFG) rickettsiosis is widely endemicthroughout the world, being known as Rocky Mountainspotted fever, Siberian tick typhus, boutonneuse fever,rickettsialpox, Australian tick typhus, and so on. In Japan,Mahara et al. (10) first reported three cases of SFG rickettsi-osis that occurred in Anan City, Tokushima, in 1984. In1986, the causative rickettsia was isolated by Uchida et al.from a patient in Kochi (13).We isolated the Katayama (8) and Abe strains in Toku-

shima in 1987 and 1988, respectively, and the Misaka strainin Hyogo in 1988 from patients with SFG rickettsiosis inJapan. In the present paper, we report the characteristics ofthese strains from both biological and immunological view-points.

MATERIALS AND METHODS

Rickettsiae. The SFG rickettsiae used are shown in Table1. The Katayama strain was isolated from a 70-year-oldwoman, a farmer, in Anan City, Tokushima, in 1987 (8). TheMisaka strain was isolated from a 68-year-old man, a for-ester, on Awaji Island, Hyogo, in 1988. The Abe strain wasisolated from a 77-year-old woman, also a farmer, in AnanCity, Tokushima, in 1988. These strains were isolated byinoculation of peripheral blood into nude mice. Rickettsiarickettsii Smith, R. conorii Moroccan, R. akari MK (Kap-lan), and R. montana tick strain were obtained from theAmerican Type Culture Collection, Rockville, Md. R. sibir-ica 246 and Rickettsia sp. strain Thai TT-118 were obtainedfrom N. Tachibana (Miyazaki Medical college, Miyazaki,Japan). R. australis Phillips was obtained from N. D. Stall-man (Laboratory of Microbiology and Pathology, Brisbane,Queensland, Australia). All rickettsial strains were passagedin mice or cultured cells (BSC-1 cells, Vero cells, or L cells)in our laboratory.

Animals. Congenitally athymic nude mice of BALB/cbackground and BALB/c mice were obtained from our

* Corresponding author.

breeding colony. Cyclophosphamide-treated BALB/c micewere inoculated subcutaneously with 2.5 to 5.0 mg of thedrug at the same time as they were inoculated with rickett-siae (7). Adult male guinea pigs weighing 400 to 650 g wereobtained commercially.

Preparation of immune mouse sera. BALB/c mice (8 to 12weeks old) were inoculated intraperitoneally with 0.2 ml ofpH 7.2 phosphate-buffered saline (PBS) containing a 10%spleen suspension from nude mice infected with each strain.About 6 weeks later, the mice were exsanguinated, and thesera were pooled and stored frozen at -20°C (5).

Production of monoclonal antibodies. Anti-Katayamamonoclonal antibodies were produced by the method de-scribed previously (6). Briefly, 6-week-old female BALB/cmice were inoculated intraperitoneally with 0.2 ml of PBScontaining a 10% spleen suspension from a nude mouseinfected with the Katayama strain. The mice received abooster immunization in the same way about 6 to 8 weekslater. Three days later, spleen cells were harvested and fusedwith P3X63Ag8.653 mouse myeloma cells by use of 45%polyethylene glycol 4000 (Wako Pure Chemical Industries,Osaka, Japan) at a spleen cell/myeloma cell ratio of 10:1. Thecells were distributed into 96-well microplates in HATselective medium; HAT medium consisted of RPMI 1640with 20% fetal calf serum, 2 mM L-glutamine, 10-4 Mhypoxanthine, 4 x 10-7 M aminopterin, 1.6 x 10-5 Mthymidine, streptomycin (100 ,ug/ml), and penicillin G (100U/ml). When clones appeared macroscopically between 10and 20 days, the supernatant fluids were screened by anindirect immunofluorescence assay (IFA). Anti-Katayamaantibody-secreting hybridomas were cloned twice or moreby limiting dilution. After the cloning, the antibody-pro-ducing hybridomas were inoculated into BALB/c mice pre-viously treated with pristane (2,6,10,14-tetramethylpentade-cane; Aldrich Chemical Co., Milwaukee, Wis.). After about10 to 14 days, ascitic fluid was harvested and sedimented bycentrifugation to remove cells and fibrin clots, and the clearascitic fluid was stored frozen at -20°C.

Determination of antibody class and subclass. Cell culturesupernatants containing monoclonal antibodies were con-

887

on October 31, 2020 by guest

http://iai.asm.org/

Dow

nloaded from

888 OKADA ET AL.

TABLE 1. Strains of spotted fever group rickettsiae used in this study

Strain Host Geographical source Yr Passage"

Katayama Human Japan 1987Misaka Human Japan 1988Abe Human Japan 1988R. rickettsii Smith Human Montana 1946 GP/2, CE(YS)/18R. sibirica 246 Dermacentor nuttalli USSR 1945 UnknownR. conorii Moroccan Unknown Morocco 1953 GP/numerous, CE/292, Vero/5, CE(YS)/4R. akari MK (Kaplan) Human New York 1946 M/2, CE/12, GP/1, CE/3R. australis Phillips Human Australia 1944 Unknown, BGM/1R. montana tick D. andersoni and D. variabilis Montana 1963 CE/20Thai TT-118 Ixodes and Rhipicephalus larval ticks Thailand 1962 Unknown

a GP, Guinea pig; CE, chicken embryo; YS, yolk sac; Vero, Vero cells; M, mouse; BGM, BGM cells. Numerals (or description) after the shills indicate thepassage numbers.

centrated to about 1:50 by the 50% saturated ammoniumsulfate precipitation method and analyzed by the micro-Ouchterlony method against rabbit antisera reactive withmouse-specific immunoglobulin G (IgG), IgGl, IgG2a,IgG2b, IgG3, IgM, and IgA (kappa and lambda chains; MilesScientific, Div. Miles Laboratories, Inc., Naperville, Ill.).

Reactivity of monoclonal antibodies and immune sera. Thereactivity of monoclonal antibodies was examined by theIFA by use of a previously described method (4). Briefly,antigens for the IFA were prepared from infected BSC-1cells, Vero cells, and L cells in tissue culture. The cellsinfected with the rickettsiae were harvested, destroyed witha Dounce homogenizer, and centrifuged at 150 x g for 5 min.The supernatant fluid was aspirated and centrifuged at13,000 x g for 30 min. The pellet was suspended in PBS,applied to microscope slides, air dried, and fixed in acetone.Ascitic fluids containing monoclonal antibodies were used asantibody samples. The initial antibody dilution was 1:10 inPBS, and subsequent serial twofold dilutions were prepared.Each antibody dilution was overlaid on the rickettsial anti-gens. The slides were incubated in a moist chamber for 30min at 37°C, washed in PBS, and air dried. Fluorescein-conjugated goat anti-mouse immunoglobulins (IgA, IgG, andIgM [heavy- and light-chain specific; Organon Teknika,Malvern, Pa.) were overlaid on the rickettsial antigens, andthe slides were incubated, washed, and air dried as describedabove. The slides were finally examined with a UV micro-scope (Universal type; Zeiss, Oberkochen, Federal Republicof Germany). Fluorescence was graded from 4+ to negativeaccording to its intensity. The antibody titer was consideredpositive if the intensity was 3+ or more. The endpoint of theantibody reaction was defined as the highest positive dilu-tion.The reactivity of immune mouse sera was examined in the

same way. In the case of immune guinea pig sera, fluores-cein-conjugated goat anti-guinea pig IgG (heavy- and light-chain specific; Organon Teknika) was used as the secondserum.Smears prepared from the peritoneum of infected mice

were also used as antigens.

RESULTS

Biological characteristics of the causative agent of SFGrickettsiosis in Japan. The pathogenicity of the isolates wasexamined to reveal their biological characteristics in exper-imental animals. Most of the nude mice inoculated intraper-itoneally with 0.2 to 0.5 ml of a 10% spleen suspension from

a nude mouse infected with the Katayama, Misaka, or Abestrain showed weakness and splenomegaly and died within 2or 3 weeks after infection. The causative rickettsiae weredemonstrated in smears prepared from the peritoneum of themice by Giemsa staining and the IFA. Normal BALB/c miceshowed ruffled fur between 5 and 10 days after infection andthen recovered from the disease and acquired immunity, asproved by elevation of the antibody to the homologous strainand other SFG rickettsiae revealed by the IFA. All of thecyclophosphamide-treated mice inoculated with the Kata-yama, Misaka, and Abe strains died between 4 and 7 daysafter infection, and the causative rickettsiae were demon-strated as in infected nude mice.Adult male guinea pigs inoculated intraperitoneally with 3

ml of a 10% spleen suspension from a nude mouse or acyclophosphamide-treated mouse infected with the Kata-yama or Misaka strain had fever and showed swelling andredness of the scrotum between 2 and 8 days after infection.However, no necrosis of the scrotum, ears, or footpads wasobserved. They then recovered from the disease and ac-quired immunity, as proved by elevation of the antibody tothe corresponding strain revealed by the IFA. The temper-ature curve of a guinea pig infected with the Katayamastrain is shown in Fig. 1. In this case, fever was defined as39.6°C or more from measurement of the rectal temperaturewith an electronic clinical thermometer (Terumo, Tokyo,Japan). The serum obtained from this guinea pig on day 44after infection showed elevation of the antibody to thehomologous strain. The scrotal reaction of this guinea pig atday 7 after infection is shown in Fig. 2 alongside a normalcontrol.The Katayama, Misaka, and Abe strains propagated well

in Vero cells cultured in Eagle minimum essential medium(Nissui Pharmaceutical Co., Tokyo, Japan) containing 2%fetal calf serum without antibiotics when a 10% spleensuspension from an infected nude mouse or a cyclophospha-mide-treated mouse was used as the inoculum. The rickett-sial particles appeared as diplobacillary and diplococcalforms, similar to other SFG rickettsiae.

Serological characteristics of the causative agent of SFGrickettsiosis in Japan. The serological characteristics of theKatayama and Misaka strains were analyzed by the cross-immunofluorescence antibody method. Strains Katayama,Misaka, R. rickettsii Smith, R. sibirica 246, R. conoriiMoroccan, R. akari MK (Kaplan), and R. australis Phillips,R. montana tick strain, and strain Thai TT-118 were used asantigens. Samples of immune mouse serum against strainsKatayama, Misaka, R. sibirica 246, R. australis Phillips, and

INFECT. IMMUN.

on October 31, 2020 by guest

http://iai.asm.org/

Dow

nloaded from

SPOTTED FEVER GROUP RICKETTSIAE IN JAPAN 889

Days 0 5 10 15 20 25 30 35 40

(4C)1 10% spleen suspension of nude mouse4infected with the Katayama strain in PBS. 3ml i.p. exsanguinated

4011At39

T

38

37-

36

35

Scrotalreaction

IFA titer<10 160

FIG. 1. Temperature (T) curve of a Katayama-infected guinea pig. i.p., Intraperitoneally.

Thai TT-118 were used as antisera (Table 2). The antiserumagainst the Katayama strain reacted at a high titer with thehomologous antigen and the Misaka antigen. This antiserumalso reacted with other SFG rickettsiae at relatively lowtiters. The antiserum against the Misaka strain reacted at ahigher titer with the homologous antigen and the Katayamaantigen than with the other SFG rickettsiae. The antiseraagainst R. sibirica 246 and strain Thai TT-118 reacted at arelatively low titer with the respective homologous strains.From these results, it was difficult to draw a definite conclu-sion about the serological characteristics of the isolates.To reveal the immunological properties of the isolates in

detail, we established 21 hybridomas secreting monoclonalantibodies against the Katayama strain. These were desig-nated KMA 1 to 21. The isotypes of the monoclonal anti-bodies were determined by the micro-Ouchterlony method.They were classified into 3 clones of IgGl, 11 clones ofIgG2a, and 7 clones of IgG3 (Table 3). Light chains of allclones were of the kappa type.The reactivities of the anti-Katayama monoclonal antibod-

ies with the homologous strain and representative strains ofSFG rickettsiae determined by the IFA are shown in Table 4.KMA 1 to 5 reacted with the Katayama strain specifically. Itwas ascertained that the Katayama strain has a strain-specific antigen(s) which is different from those of the otherSFG rickettsiae. KMA 15 to 21 reacted with all strains ofSFG rickettsiae. It was thus shown that the Katayama strainshares a common antigen(s) with the strains of the other SFGrickettsiae. KMA 6 to 14 reacted with one or two strains ofSFG rickettsiae besides the Katayama strain and at anintermediate level. KMA 6, 7, 8, 9, 13, and 14 reacted withstrain Thai TT-118. KMA 10 reacted with the R. montanatick strain. KMA 11 and 13 reacted with R. rickettsii Smith,and KMA 12 and 14 reacted with R. sibirica 246. On theother hand, none of the monoclonal antibodies reacted withR. conorii Moroccan, R. akari MK (Kaplan), or R. australis

Phillips, except for those antibodies that reacted with allstrains of SFG rickettsiae.

Furthermore, the serological characteristics of the Misakaand Abe strains were analyzed with the strain-specific anti-bodies to the Katayama strain, KMA 1 to 5. All of theseantibodies reacted with the Misaka and Abe strains (Table5). Thus, it was apparent that the Misaka and Abe strainsbelong to the Katayama strain serotype.

DISCUSSION

It is well known that SFG rickettsiosis is widely endemicthroughout the world. However, until recently, only scrubtyphus was thought to be an important rickettsial disease inJapan. In 1984 and thereafter, patients with SFG rickettsio-sis were reported in various parts of Japan, includingTokushima, Kochi, Miyazaki, and Hyogo (10, 12, 14, 19).The causative rickettsia, showing cross-reactivity with R.montana, was isolated by Uchida et al. from a patient inKochi in 1986 (13), and subsequently four strains wereisolated from patients who were ill in 1985 and 1986 (15). Inthe present study, we attempted to reveal the biological andimmunological characteristics of the causative agents ofSFG rickettsiosis in Japan which we had isolated frompatients in Tokushima and Hyogo.

It has been shown that SFG rickettsiae, especially R. akariand R. australis, are pathogenic for adult white mice (17). Inthe present study, it was revealed that the causative agentsof SFG rickettsiosis in Japan were also moderately patho-genic for normal BALB/c mice. In addition, their pathoge-nicity for nude mice and cyclophosphamide-treated micewas striking. This result suggested that a sound immuneresponse is important for recovery from the disease.The isolates caused fever, a scrotal reaction, and serocon-

version in infected guinea pigs. This finding corresponds to

VOL. 58, 1990

on October 31, 2020 by guest

http://iai.asm.org/

Dow

nloaded from

890 OKADA ET AL.

TABLE 2. Cross-immunofluorescence antibody method resultsfor spotted fever group rickettsiae tested with immune mouse sera

Titer' of the following immune serum:

Antigen ThKatayama Misaka R. sibirica R. australis TTh118

Katayama 2,560 1,280 160 80 160Misaka 1,280 2,560 160 80 160R. rickettsii 160 320 80 80 160

SmithR. sibirica 80 320 320 40 80

246R. conorii 160 320 160 40 160Moroccan

R. akari MK 160 160 80 160 80(Kaplan)

R. australis 80 80 80 1,280 80Phillips

R. montana 160 160 80 80 80tick

Thai TT-118 80 320 80 40 160

' Expressed as the reciprocal of the dilution.

p..4. Wwom~~~~~.: .:..

_X w ,-MMW)L- ,

FIG. 2. (A) Scrotal reaction of a Katayama-infected guinea pigon day 7. (B) Normal control.

the pathogenicity of other SFG rickettsiae for guinea pigs(18); thus, the present isolates are not distinguishable fromother SFG rickettsiae from this viewpoint. However, itseems that the causative agents of Japanese SFG rickettsio-sis are less virulent for guinea pigs than are R. rickettsii R-type strains, which evoke necrotic lesions of the scrotum,footpads, and ears and invariably kill the animals (1).From a serological analysis with immune mouse sera by

the cross-immunofluorescence antibody method, it becameapparent that the isolates may be closely related to eachother serologically. However, it was difficult to reveal theirimmunological relationships in detail because of wide cross-reactivities among SFG rickettsiae. In the present study, weproduced 21 monoclonal antibodies against a new isolate,

the Katayama strain, from a patient with SFG rickettsiosis inJapan and analyzed their serological properties.Seven anti-Katayama monoclonal antibodies, KMA 15 to

21, reacted with all representative strains of SFG rickettsiaeused in the present study, and it was revealed that theKatayama strain has an antigen(s) in common with those ofSFG rickettsiae. Five anti-Katayama monoclonal antibod-ies, KMA 1 to 5, reacted specifically with the Katayamastrain, and it was proved directly that the Katayama strainhas a serotype-specific antigen(s) different from those of theother SFG rickettsiae used in this study. Moreover, theseserotype-specific monoclonal antibodies reacted with twoother new isolates in Japan, the Misaka and Abe strains.Therefore, it appears that these two strains belong to a newserotype, Katayama, of the spotted fever group and thatJapanese SFG rickettsiosis is a new, independent disease. Inthis regard, Uchida et al. also characterized Japanese strainsby using mouse antisera and monoclonal antibodies to R.conorii, R. rickettsii, R. sibirica, and R. akari (16).Lackman et al. (9) and Robertson and Wisseman (11) did

a classical subgrouping of SFG rickettsiae by using thecomplement fixation test and the mouse toxin neutralizationtest, respectively. Subgroup A consists of R. rickettsii andR. sibirica. R. conorii and R. parkeri are in subgroup B. R.akari and R. australis are in subgroup C, and R. montanaand Western Montana U are in subgroup D. Pakistan JC-880

TABLE 3. Isotypes of anti-Katayama monoclonal antibodies

KMAa Isotypeb KMA Isotype

1 IgG2a 12 IgG2a2 IgGl 13 IgG2a3 IgGl 14 IgG2a4 IgG2a 15 IgG35 IgGl 16 IgG36 IgG2a 17 IgG37 IgG2a 18 IgG38 IgG2a 19 IgG39 lgG2a 20 IgG310 IgG2a 21 IgG311 IgG2a

aKMA, Anti-Katayama monoclonal antibody.b Light chains of all clones were of the kappa type.

INFECT. IMMUN.

on October 31, 2020 by guest

http://iai.asm.org/

Dow

nloaded from

SPOTTED FEVER GROUP RICKETTSIAE IN JAPAN 891

TABLE 4. Reactivity of anti-Katayama monoclonal antibodies with spotted fever group rickettsiae

Titerf of the following KMA"Antigen

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21

Katayama 20,480 2,560 1,280 2,560 1,280 2,560 2.560 2,560 1,280 640 80 2,560 5,120 40 640 320 160 160 160 80 80R. rickettsii --.40- 2,560 -320 320 320 160 160 160 40

SmithR. sibirica246 -.2,560 - 20 640 160 320 320 160 320 160R. conorii - ..... 320 160 320 320 160 320 160Moroccan

R. akari MK -...640 320 640 640 320 320 160(Kaplan)

R. aiustr-alis -.-.640 320 320 320 320 320 160Phillips

R. montana - 640 -.320 160 320 80 160 160 80tick

Thai TT-118 - - - - - 2,560 1,280 320 640 - - - 2,560 20 640 160 160 160 160 160 80

" Expressed as the reciprocal of the dilution.b KMA, Anti-Katayama monoclonal antibody.'-, Negative or <1:10.

and Thai TT-118 belong to subgroup E. In the present study,we compared the serological characteristics of the isolates inJapan and representative strains of these SFG rickettsiasubgroups. Six of nine monoclonal antibodies with interme-diate reactivity reacted with strain Thai TT-118. Moreover,two of these antibodies reacted with R. ri(kettsii and R.sibirica. KMA 13 showed cross-reactivities with Katayama,Thai TT-118, and R. rickettsii Smith. KMA 14 showedcross-reactivities with Katayama, Thai TT-118, and R. sibir-ica 246. On the other hand, the nine antibodies did not reactwith R. conorii, R. akari, or R. alustralis. These findingssuggested that the Katayama strain may be closely relatedto the Thai TT-118 strain, i.e., subgroup E, serologically,a possibility which is of interest because these strainsare both endemic in east Asia. It was also suggested thatthe Katayama strain may be more closely related to R.rickettsii and R. sibirica (subgroup A) than to R. (C'onorii(subgroup B), R. akari (subgroup C), and R. aiustralis(subgroup C).

In east Asia, it has been reported that R. akar-i exists in theRepublic of Korea (3) and that R. sibiri(a is distributed innorthern China (2). Thus, the distribution of SFG rickettsiaeis complicated and diversified. Hereafter, it will be importantto reveal the distribution of this new rickettsia serotype ofthe spotted fever group for etiological, epidemiological, and

TABLE 5. Reactivity of the strain-specific anti-Katayamamonoclonal antibodies with spotted fever group rickettsiae

Titer" of the following KMA":Antigen

1 2 3 4 5

Katayama 20,480 2,560 1,280 2,560 1,280Misaka 10,240 1,280 1,280 2,560 640Abe 10,240 2,560 640 1,280 640R. rickettsii SmithR. sibiricCa 246R. conorii MoroccanR. akari MK (Kaplan)R. aiustrcalis PhillipsR. montana tickThai TT-118

Expressed as the reciprocal of the dilution.KMA, Anti-Katayama monoclonal antibody.-. Negative or <1:10.

clinical studies of the disease. For this purpose, Katayamaserotype-specific monoclonal antibodies KMA 1 to 5 will beextremely useful for identifying the causative agent.

ACKNOWLEDGMENTS

We thank F. Mahara and K. Kodama for giving us informationabout clinical cases, N. Tachibana for supplying R. sibiric a and ThaiTT-118, and N. D. Stallman for supplying R. austr-alis.

LITERATURE CITED1. Elisberg, B. L., and F. M. Bozeman. 1979. The rickettsiae, p.

1061-1108. In E. H. Lennette and N. J. Schmidt (ed.), Diagnos-tic procedures for viral, rickettsial and chlamydial infections,5th ed. American Public Health Association, Inc., Washington,D.C.

2. Fan, M. Y., X. J. Yu, and D. H. Walker. 1988. Antigenicanalysis of Chinese strains of spotted fever group rickettsiae byprotein immunoblotting. Am. J. Trop. Med. Hyg. 39:497-501.

3. Jackson, E. B., J. X. Dnauskas, M. C. Coale, and J. E. Smadel.1957. Recovery of Rickettsia akairi from the Korean voleMicrotus fortis pelliceuts. Am. J. Hyg. 66:301-308.

4. Kanemitsu, N. 1987. Production and characterization of themonoclonal antibodies to Rickettsia tsiwtsiuga;nilushi. Kansen-shogaku Zasshi 61:819-829.

5. Kobayashi, Y. 1969. Analytical serology of rickettsiae, p. 203-256. In J. B. G. Kwapinski (ed.), Analytical serology of micro-organisms, vol. 1. John Wiley & Sons, Inc., New York.

6. Kobayashi, Y., H. Hasegawa, T. Oyama, T. Tamai, and T.Kusaba. 1984. Antigenic analysis of Japanese encephalitis virusby using monoclonal antibodies. Infect. Immun. 44:117-123.

7. Kobayashi, Y., N. Tachibana, I. Matsumoto, T. Oyama, and T.Kageyama. 1978. Isolation of very low virulent strain of Rick-ettsia tsiatsuigaicmnushi by use of cyclophosphamide-treated mice,p. 181-188. In J. Kazar, R. A. Ormsbee, and I. N. Tarasevich(ed.), Rickettsiae and rickettsial diseases. Veda, PublishingHouse of the Slovak Academy of Sciences, Bratislava, Czech-oslovakia.

8. Kobayashi, Y., Y. Tange, N. Kanemitsu, T. Okada, and F.Mahara. 1988. The causative agent from a patient with spottedfever group rickettsiosis in Tokushima, Japan. KansenshogakuZasshi 62:1132-1137.

9. Lackman, D. B., E. J. Bell, H. G. Stoenner, and E. G. Pickens.1965. The Rocky Mountain spotted fever group of rickettsias.Health Lab. Sci. 2:135-141.

10. Mahara, F., K. Koga, S. Sawada, T. Taniguchi, F. Shigemi, T.Suto, Y. Tsuboi, A. Ooya, H. Koyama, T. Uchiyama, and T.Uchida. 1985. The first report of the rickettsial infections ofspotted fever group in Japan; three clinical cases. Kansenshog-

VOL. 58, 1990

on October 31, 2020 by guest

http://iai.asm.org/

Dow

nloaded from

892 OKADA ET AL. INFECT. IMMUN.

aku Zasshi 59:1165-1172.11. Robertson, R. G., and C. L. Wisseman, Jr. 1973. Tick-borne

rickettsiae of the spotted fever group in West Pakistan. II.Serological classification of isolates from West Pakistan andThailand: evidence for two new species. Am. J. Epidemiol.97:55-64.

12. Tachibana, N., E. Shishime, A. Okayama, J. Ishizaki, K. Murai,S. Shioiri, K. Tsuda, and T. Oshikawa. 1987. Two cases ofspotted fever rickettsiosis in Kyushu. Kansenshogaku Zasshi61:1166-1172.

13. Uchida, T., F. Tashiro, T. Funato, and Y. Kitamura. 1986.Isolation of a spotted fever group rickettsia from a patient withfebrile exanthematous illness in Shikoku, Japan. Microbiol.Immunol. 30:1323-1326.

14. Uchida, T., Y. Tsuboi, A. Oya, T. Funato, and Y. Kitamura.1986. Serologic studies of spotted fever group rickettsiosisoccurred in Muroto-City, Kochi Prefecture. KansenshogakuZasshi 60:141-144.

15. Uchida, T., T. Uchiyama, and A. H. Koyama. 1988. Isolation ofspotted fever group rickettsiae from humans in Japan. J. Infect.Dis. 158:664-665.

16. Uchida, T., X. Yu, T. Uchiyama, and D. H. Walker. 1989.Identification of a unique spotted fever group rickettsia fromhumans in Japan. J. Infect. Dis. 159:1122-1126.

17. Walker, D. H., and M. G. Peacock. 1988. Laboratory diagnosisof rickettsial diseases, p. 135-155. In D. H. Walker (ed.),Biology of rickettsial diseases, vol. II. CRC Press, Inc., BocaRaton, Fla.

18. Woodward, T. E., and E. B. Jackson. 1965. Spotted feverrickettsiae, p. 1095-1129. In F. L. Horsfall, Jr., and I. Tamm(ed.), Viral and rickettsial infections of man, 4th ed. J. B.Lippincott Co., Philadelphia.

19. Yamamoto, S., N. Kawabata, T. Uchiyama, and T. Uchida. 1987.Evidence for infection caused by spotted fever group rickettsiain Kyushu, Japan. Jpn. J. Med. Sci. Biol. 40:75-78.

on October 31, 2020 by guest

http://iai.asm.org/

Dow

nloaded from