Jap. J. M. Sc.& Biol., 17, 59-72, 1964

STRAIN VARIATION OF RICKETTSIA ORIENTALIS IN THE

COMPLEMENT FIXATION TEST *

AKIRA SHISHIDO

Department of Virology & Rickettsiology, the National Institute of Health, Tokyo

(Received : Jannuary 16th, 1964)

It is well known that Tsutsugamushi disease or scrub typhus in Japan has been

prevalent in a limited northern district of the country, but Japanese investigators were not aware that there was a different type of scrub typhus in other parts of Japan, although some sporadic cases had been reported in the foothills of Mt. Fuji (Sakita, 1934), and Kanagawa Prefecture (Fukuzumi et al., 1950), which however were not reconfirmed by other investigators before 1948 when an outbreak of this disease among American soldiers of the Occupation Forces during maneuvers in the foothills of Mt. Fuji proved its existence in a district other than the above-stated well-known endemic foci of the country (406 MGL, 1948).

Shortly afterward, organisms belonging to the Rickettsia orientalis group were isolated by several Japanese investigators from patients, rodents and trombiculid mites in different

parts of the country (Dept. Health, Tokyo, 1952, Kawamura et al., 1952, Kasahara et al, 1953, Kimura et al., 1953, Kitaoka et al., 1953 a, Kitaoka et al., 1953 b, Ogasawara et al., 1953). Thereafter, we Japanese investigators generally considered these newly discovered diseases to be a new type of scrub typhus differing from the well-known classic type of the disease. Then, in 1954, a study team on scrub typhus was organized by a number of Japanese investigators including the author, headed by the late Dr. Takeo Tamiya, and an epidemiological study on scrub typhus of both the types has been carried out by this team for several years. Through an epidemiological survey carried out by this team, a) a large number of field rodents trapped throughout Japan were found very often to harvor this organism, b) the vector mites of this rickettsia were disclosed to be consisting of a number of different species, occurring at various seasons, c) the severity of the patients' symptoms seemed to vary according to the locality (Tamiya, 1962).

The author's work in this team was concentrated on the classification of strains from different sources from the standpoint of antigenicity or immunological relationships, because it seemed very desirable for us to determine whether or not any relation exsists between the serological type of strain and its sources (Shishido et al., 1961, Shishido et al., 1962a, Shishido, 1962).

The present study deals with variation of Rickettsia orientalis in the complement

fixation test and its relationship to the sources. As is well known a heterogeneity exsists

* This paper was read at the 7th International Congress of Tropical Medicine and Malaria held

at Rio de Janeiro on September 1-10, 1963.宍戸 亮(国 立予防衛生研究 所 ウィルス リケッチア部)

59

60 SHISHIDO Vol. 17

among strains of scrub typhus in the complement fixation test (Bengtson, 1945). It was attempted to classify the strains isolated from different sources in Japan into some types by the complement fixation test, and we succeeded in classifying those strains into three types as will be indicated in this paper.

MATERIALS AND METHODS

The procedure of preparation of complement fixing antigen which was employed in this ex-periment relied principally upon differential centrifugation that was used by several workers in studies on rickettsial metabolism (Bovarnick et al., 1950, Wisseman et al., 1951, Yamamoto et al., 1958, Kohno, 1959). An outline of the procedure is illustrated in Fig. 1. As is shown in the figure, the yolk sac infected with rickettsiae was treated with ion-exchange resin XE-64 and bovine serum albumin in a differential centrifugation. UV irradiation was conducted for inacti-vation of thus purified rickettsiae. Therefore, it can be safely said that the antigen thus prepared was composed of partially purified inactive rickettsial particles in contrast to a soluble antigen which is generally used for CF antigen of Rickettsia orientalis (Bengtson, 1945, Topping et al., 1946, Wolf e et al., 1946).

Antiserum of each strain tested was obtained by infecting five or more guinea pigs intracere-brally with 0.1 cc of a 10 % suspension of spleen-liver homogenate of mice infected with each strain of Rickettsia orientalis. After 4 weeks, in the convalescent stage, the serums were collected from the animals thus immunized, and the pooled serum was used for test (Jackson et al., 1957).

In the complement fixation test, an attempt was made to determine whether or not the disposo-tray (Ace Scientific Co., Linden, N. J., U. S. A.) method for CFT, which has been

Table 1. The influence of temperature in storage on the titers of CF antigen

-* Not done

1964 VARIATION OF R. ORIENTALIS

61

* XE-64 : Amberlite (Rohm & Haas Co.).** sucrose PGy : sucrose P (sucrose PG minus glutamate) + 1 % glycerin.

*** BSA : bovine serum albumin fraction V

Armour Co. U. S. A.

Fig. 1. Preparation of CF antigen

62 SHISHIDO Vol. 17

already applied for several viral diseases (Okuno et al ., 1961), was likewise useful for scrub typhus work. The results obtained with this method were reproducible and well comparable with those obtained by the usual test-tube method. Therefore , all tests thereafter were carried out by this method.

By using this method, in the complement fixation test of antiserums of a number of strains , checker-board titration was employed for comparison of antigenic differences among strains. In this case, antigen titers of different types of antigens were expressed as units to antiserums of

prototype strains.As to the stability of the antigen, it was disclosed that the titer was very stable under a

storage at 4•Ž in a refrigerator and at -20•Ž in a deep freezer for several weeks (Table 1).

It was also stable against the procedure of freez-thawing (Table 2) . Hence, the antigen was

usually stored at -20•Ž in deep freezer, and thawed at need for test.

Table 2. The influence of freezing and thawing on the titers of CF antigens

EXPERIMENTALS

Heterogeneity among Strains of Rickettsia orientalis in the Complement Fixation Test

The heterogeneity among strains of Rickettsia orientalis in the complement fixatiore test had been disclosed by Bengtson (Bengtson, 1945). At first, such a heterogeneity was examined with our particle antigens. Antigens were prepared from three egg-adapted strains, Kato, Karp and Gilliam by the method mentioned above, among which Kate strain is a reference strain of Rickettsia orientalis in this laboratory (Shishido et al., 1958), and the latter two were found to be heterotypic to each other by Bengtson . These antigens were tested against strain-specific guinea pig anti-serums of the three strains. The results of a checker-board titration among those strains indicated that each strain is composed not only of its own main specific antigen but of a small amount of antigen common to them, as given in Table 3.

In the next experiment, the CF antibody development was followed up in serums of

guinea pigs four or more times with two week intervals after inoculation with Kato, Karp, or Gilliam strain of Rickettsia orientalis. As shown in Figs. 2, 3, and 4, two

groups of animals, A and B, were used in each, and group A animals were inoculated intracerebrally as indicated before without a booster injection, but group B were injected

1964 VARIATION OF R. ORIENTALIS 63

Table 3. Different pattern of checker-board CFT shown by Kato,

Karp and Gilliam strains of R. orientalis

-at first intracerebrally as in the group A and then three times intraperitoneally each with

2.0 cc of the same material at two week intervals after the initial inoculation .However, in any case, the intraperitoneal booster injection did neither enhance the

CF antibody level beyond that caused by a single intracerebral inoculation into guinea

pigs nor modify their own antibody pattern. This means that a single intracerebral infection with each strain of Rickettsia orientalis to guinea pigs seems to be enough to obtain strain-specific antiserum from those animals for serological analysis .

Serological Classification of Strains from Different Sources

The countrywide survey on rickettsia harvoring animals carried out by Dr . Tamiya's study team revealed that rickettsia positive rodents were found not only in the limited areas known to be endemic of scrub typhus but also widely distributed from Hokkaido Island, the northernmost to Tanegashima Island , the southernmost (Fig. 5). The total number of strains isolated from field rodents reached about 500 (Tamiya , 1962). The serological classification was at first carried out on these strains of Rickettsia orientalis . Twenty strains were selected at random from among these strains according to the locality of isolation (Fig. 5) , 6 from known endemic areas, 14 from non-endemic areas (Fig. 5), and strain-specific immune serums were prepared from guinea pigs immunized with there strains according to the above-indicated method (see Materials and Methods) .

The results of complement fixation tests on these antiserums against the prepared

antigens of the three types, Kato, Karp, and Gilliam , using checker-board titration, were as follows:

64 SHISHIDO Vol. 17

Fig. 2. Time course of antibody production in sera of guinea pigs immunized with Kato strain of R. orientalis.

Fig. 3. Time course of antibody production in sera of guinea pigs

immunized with Karp strain of R. orientalis.

1964 VARIATION OF R. ORIENTALIS 65

66 SHISHIDO Vol. 17

Fig. 5. Map of Japan.

a) one strain (AR-65 No. 25) isolated in Niigata Prefecture, where a typical classic type of scrub typhus is prevailing, belongs to the Kato type (Fig. 6, see Rodent's column), b) one strain (Oshima No. 1) isolated in Oshima Island, 2 strains (AR-58 No . 2, AR-62 No. 13) isolated at the foothills of Mt. Fuji, both endemic areas of the new type of scrub typhus, and 9 strains isolated at non-endemic areas throughout Japan except Hokkaido Island belong to the Karp type (Fig. 7, see Rodent's column), c) one strain isolated in Niigata (KR-1 No. 39) and 4 strains isolated in Hokkaido Island belong to the Gilliam type (Fig. 8, see Rodent's column).

This result indicates that all strains which were isolated from field rodents so far seem to belong to either Kato, Karp or Gilliam type of Rickettsia orientalis.

As to the rickettsia positive mites, a countrywide survey for isolation of rickettsiae from mites has not been carried out as yet, but isolation of rickettsiae has been only attempted on those mites which were found infesting field rodents captured at known endemic areas with a few exception (Hokkaido).

The infesting mites on the captured rodents were collected, devided into some groups

(10-100 mites per group) according to their species, and emultions of the mites were inoculated into mice for rickettsia isolation (Asanuma et at., 1962 a, Asanuma et at., 1962b, Asanuma, Okubo, et at., unpublished data, 1963). Fourteen of about 100 strains

1964 VARIATION OF R. ORIENTALIS 67

1. Kato Type

Fig. 6. Strain variations of R. orientalis

thus isolated from trombiculid mites were tested for serological classification.

The following results were obtained : a) 5 strains isolated from Trombicula akamushi collected in Niigata Prefecture belong to the Kato type (Fig. 6, see Mite's column), b) 2 strains (AT-103, AT-104) isolated from Trombicula pallida collected in Oshima Island, one strain (Miura-P) isolated from Trombicula pallida collected at Nobi, Kanagawa Prefecture, and one strain (AT-38 No. 10) isolated from Euchoengastia ikaoensis collected at the foothills of Mt. Fuji, all endemic areas of the new type of scrub typhus, belong to the Karp type (Fig. 7, see Mite's column), c) 4 strains (At-43 No. 5, AT-43 No. 6, KT-1 No. 1, KT-23 No. 3) from Trombicula pallida collected in Niigata Prefecture and one strain (AT-32 No. 32) from Trombicula kawamurai collected in Hokkaido belong to the Gilliam type (Fig. 8, see Mite's column).

This result also indicated that all strains isolated from mites seem to belong to either

Kato, Karp or Gilliam type.As for strains of human origin, one strain (Furuyama) from patients in Niigata, and

one strain (Uemura) from a patient at the foothills of Mt. Fuji were tested for the types they belong to. The results indicated that the former strain from Niigata belonged to the Kato, and the latter from Fuji belonged to the Karp type (Fig. 6 and 7, see Human columns).

DISCUSSION AND CONCLUSION

In the serological classification of strains of Rickettsia orientalis by the CFT presented in this paper, it should be pointed out that some strains gave much more cross-reaction compared with prototype strains of Kato, Karp, and Gilliam. However, it may be

justified to classify a number of strains into three types according to their main antigen,

68 SHISHIDO Vol. 17

if the classification were used not for the purpose of antigenic analysis of each strain but for obtaining such a general idea of epidemiological status of scrub typhus in Japan as will be described later. In this sence, it will be also justified to classify antigenicity of strains in the CFT by a one-way method, that is, by analysis of antibody pattern of strain-specific immune serum of each strain against three types of antigen without testing its antigenicity in antigen prepared from each strain.

From the foregoing, the author's present conclusion about the relationship between

1964 VARIATION OF R. ORIENTALIS 69

3. Gillian Type

Fig. 8. Strain variations of R. orientalis.

Table 4. Schema on the relation between serological type and epidemiological features

the serological type of the strain and its sources is as follows (Table 4, Fig. 6, 7 and 8).a) Kato type strains were isolated from human cases of the classic type of scrub

typhus, Trombicula akamushi, and from field rodents captured in Niigata, an endemic area of this type of the disease. This may reasonably justify our use of Kato strain, which was isolated from a patient in Niigata and adapted to the yolk sac tissue in this laboratory, as a reference strain of Rickettsia orientalis (Shishido et al., 1958). In other words, the finding that strains of this type were frequently isolated from Trombicula akamushi as well as from field rodents in Niigata will be a good indication that Kato tvne strain will represent a prevalent type of the classic scrub typhus in Japan.

b) Karp type strains were isolated from human cases of the new type of scrub

70 SHISHIDO Vol. 17

typhus, Trombicula pallida, Eushoengastia ikaoensis, and also from rodents captured throughout Japan including new endemic foci of the disease. This means that the agent causing a new type of the disease seems to be the Karp type. As to the vector of the new type of scrub typhus, Trombicula scutellaris and Trombicula pallida were already reported to be a potential one in Izu Islands (Sasa, 1954) and Chiba Prefecture (Asanuma et al., 1959) and that at Miura, Kanagawa Prefecture and Shimoda, Shizuoka Prefecture (Asanuma et al., 1960) respectively. However, no strain from Trombicula scutellaris was available to test for the type it belongs to in this experiment, therefore a serological confirmation of the vector of Trombicula scutellaris was not accomplished as yet, and the role of Eusengastia ikaoensis on the vector of the scrub typhus at Fuji remained to be determined in a further epidemiological investigation. Moreover, it was neither clarified as yet what species of trombiculid mite are taking a potential role in the distri-bution of Karp type rickettsia among the field rodents throughout Japan.

c) Gilliam type strains were isolated from Trombicula pallida in Niigata , Trombi-cula kawarnurai in Hokkaido, and rodents captured at both the places. There is no case of scrub typhus, either classic or new type, from which a Gilliam type strain was isolated. If a case can be found in Hokkaido in the future, it might be highly probable that it is caused by the Gilliam type agent, and if a case occurs by attack of Trombi-cula pallida infected with Rickettsia orientalis in Niigata , it will be caused by the Gilliam type. The question of whether Trombicula pallida or any species other than Trombicula akamushi is playing a role as the vector of the classic type of scrub typhus in Japan is under study.

The author is greatly indebted to the late Dr. Takeo Tamiya, the ex-president of the Japa-nese Medical Association for his frequent encouragement and valuable advices . He is very grateful to Drs. Masami Kitaoka, Kiyoshi Asanuma and Kaoru Okubo for their active interest and advices. The author is also indebted to all staffs of Dr . Tamiya's study team, Dept. of Virology and Rickettsiology, N. I. H., Department of Hygine (Dr. K. Toyokawa , chief), Faculty of Medicine, Tokyo University, Dept. of Bacteriology (Dr. F. Shimizu , chief), Dept. of Medical Z

oology (D. R. Kano, chief), Fuculty of Medicine , Tokyo Medical Dental University, and Research Institute for Natural Resources for their cooperation and assistance .

Thanks are also due to Mrs. Michiko Hikita for her excellent technical assistance .This work is aided in part through a grant for scientific research under Cont . No. DA-92-

557, FED-29485, 31315, 33903, 35435 between Dr . Tamiya and office of the surgeoli general, U. S. A.

REFERENCES

ASANUMA, K., KITAOKA, M., OKUBO, K., KUMADA, N., SUZUKI, M., KARASAWA, T., KUGOH, T., YAMAMOTO, H., KAWAMURA, A., MIYAMOTO, T., AKIYAMA, J., NAKAGAWA, H. & HORI, K.(1959): Evidence for Trombicula scutellaris to be a vector of scrub typhus in Chiba Prefecture, Japan. Jap. J. Sanit. Zool ., 10, 232-244 (text in Japanese).

ASANUMA, K., KANO, R., KUMADA, N., KITAOKA, M., OKUBO, K., et al.(1960): Evidences for Trombicula pallida to be a vector of scrub typhus in Miura and Shimoda districts . Jap. J. Sanit. Zool., 11, 59-60 (text in Japanese).

ASANUMA, K., OKUBO, K., KUMADA, N. & KITAOKA, M.(1962 a): Determination of the

vector mites of scrub typhus in Japan. Jap. J. Med . Sci.& Biol., 15, 297-307.ASANUMA, K., KITAOKA, M., OKUBO, K., KUMADA, N., KUGOH, T., et al.(1962 b): Determi-

nation of the vector mite of scrub typhus in southern parts of Miura and Izu Peninsula , Japan. Miscel. Rep. Res. Inst. Natural Resources, No. 58, 37-55 (text in Japanese) .

1964 VARIATION OF R. ORIENTALIS 71

BENGTSON, I. A.(1945): Apparent serological heterogeneity among strains of tsutsugamushi

disease. Pub. Health Rep., 60, 1483-1488.

BOVARNICK, M. R. & MILLER, J. C.(1950): Oxidation and transamination of glutamate by

typhus rickettsias. J. Biol. Chem., 184, 661-676.

Dept. Health, Tokyo (1952): Studies on Shichito Fever Part 1.(text in Japanese).406th Medical General Laboratory Annual Historical Report, 1948, U. S. Army (1948): A new

endemic area of Tsutsugamushi disease in Japan. 90-99.FUKUZUMI, S., OBATA, Y. & OWADA, H.(1950): Studies on Tsutsugamushi disease and

Trombicula in the vicinity of Yokohama. Trans Soc. Path. Japan, 39, 40-42 (text in Japa-nese).

JACKSON, E. B., DANAUSKAS, J. X., SMADEL, J. E., FULLER, H. S., COALE, M. C. & BOZE-

MAN, F. M.(1957): Occurrence of Rickettsia tsutsugamushi in Korea rodent and chiggers.

Amer. J. Hyg., 66, 309-320.

KASAHARA, S., ISHIDA, K., MAKING, K., SUZUKI, T., TSUKAKOSHI, N., HOSOYA, Y., SEKIYA, S., HASHIMOTO, S., KO, T., ISOYAMA, K. & ISHII, T.(1953): Studies on rickettsiosis in Chiba Prefecture. Outline of so-called Twenty-day fever. Rep. Hyg. Lab., Chiba Pref., 3-5

(text in Japanese).KAWAMURA, A., SUZUKI, K., ODA, A., IWASAKI, T., SASA, M., NAKAYA, I., OHARA, R.,

KAWASHIMA, F. & KAWASHIMA, T.(1952): Studies on Tsutsugamushi disease in Kochi Prefecture. Tokyo Med. J., 69, 495-498 (text in Japanese).

KIMURA, R. & TOKUDA, M.(1953): Studies on the causative agent and mode of transmission of Tango-netsu. Rept. Co-oper. works sponsered by Ministry of Education, 603 (text in Japa-nese).

KITAOKA, M., OKUBO, K., ASANUMA, K., NOGUCHI, M.& MOCHIZUKI, H.(1953 a): Studies on scrub typhus on Izu Pennisula. Rep. Izu Red Cross Hospital, No. 1, 10-11 (text in Japa-nese).

KITAOKA, M., SHISHIDO, A., TANIGUCHI, H. & ASANUMA, K.(1952 b): Studies on scrub typhus in Kanagawa Prefecture. Rep. Hyg. Lab., Kanagawa Pref., 13-19 (text in Japanese).

KOHNO, S.(1959): Studies on metabolism of rickettsiae. I. Studies on dehydrogeneses of

Rickettria mooseri. Jap. J. Med. Sci. & Biol., 12, 375-385.

OGASAWARA, H., ANDO, K., IIZUKA, M., NOGUCHI, M., MOCHIZUKI, H., NOBEKI, M.& SAKAI, S.(1953): Studies on Tsutsugamushi disease on Izu Peninsula. Tsutsugamushi disease like disease at the area around Izu Shuzenji. Rep. Izu Red Cross Hospital, No. 1, 5-8 (text in Japanese).

OKUNO, T., OYA, A. & ITO, T.(1961): The identification of Negishi virus. A presumely

new member of Russian spring-summer encephalitis virus family isolated in Japan. Jap. J.

Med. Sci. & Biol., 14, 51-59.

SAKITA, H.(1934): A case of suspected Tsutsugamushi disease encountered at the foothills of Mt. Fuji. Japan Army Med., No. 248, 289 (text in Japanese).

SASA, M.(1954): Comparative epidemiology of tsutsugamushi disease in Japan. Jap. J. Exper.

Med., 24, 335-361.

SHISHIDO, A., OHTAWARA, M., TATENO, S., MIZUNO, S., OGURA, M. & KITAOKA, M.(1958): The nature of immunity against scrub typhus in mice. I. The resistance of mice, surviving subcutaneous infection of scrub typhus rickettsia, to intraperitoneal reinfection of the same agent. Jap. J. Med. Sci.& Biol., 15, 308-322.

SHISHIDO, A., KOHNO, S., HIKITA, M. & KITAOKA, M.(1961): Studies on the complement fixation test of Tsutsugamushi disease. 1. Purification of antigen and its specificity. Virus, 11, 246-247 (text in Japanese).

SHISHIDO, A., HIKITA, M., HOSHINO, A., MIYAMOTO, K., & KITAOKA, M.(1962): Studies on the complement fixation test of Tsutsugamushi disease. 2. Serological classification of newly isolated strains of Rickettsia orientalis in Japan by micro-CFT. Virus, 12 148-149

(text in Japanese).

72 SHISHIDO Vol. 17

SHISHIDO, A.(1962): Identification and serological classification of the causative agent of scrub

typhus in Japan. Jap. J. Med. Sci. & Biol., 15, 308-322.

TAMIYA, T.(1962): Recent advances in studies of Tsutsugamushi disease in Japan. Medical

Culture Inc., Tokyo.

TOPPING, N. H. & SHEPARD, C. C.(1946): A method for the preparation tsutsugamushi

(scrub typhus) antigen from infected yolk sacs. Pub. Health Pep., 61, 778-781.WISSEMAN, C. L., JACKSON, E. B., HAHN, F. E., LEY, A. C. & SMADEL, J. E.(1951):

Metabolic studies of rickettsiae. I. The effect of antimicrobial substance and enzyme inhibi-

tors on the oxidation of glutamate by purified rickettsiae. J. Immunology, 67, 123-136.

WOLFS, D. M., VAN DER SCHEER, J., CLANCY, C. F. & COX, H. R. 1946): A method for the preparation of complement-fixing antigens in a study of experimental tsutsugamushi disease (scrub typhus). J. Bact., 51, 247-255.

YAMAMOTO, T., KAWAMURA, A., HARA, H. AIKAWA, K.(1958): Partial purification of

rickettsiae with a cation exchange resin. I. Method of purification. Jap. J. Exper. Med.,

28, 329-336.