ANTIGENIC ANALYSIS BY DIRECT IMMUNOFLUORESCENCE OF 114 ISOLATES OF RICKETTSIA TSUTSUGAMUSHI RECOVERED FROM FEBRILE PATIENTS IN RURAL MALAYSIA*

Transcription

1 Japan. J. Med. Sci. Biol., 32, , 1979 ANTIGENIC ANALYSIS BY DIRECT IMMUNOFLUORESCENCE OF 114 ISOLATES OF RICKETTSIA TSUTSUGAMUSHI RECOVERED FROM FEBRILE PATIENTS IN RURAL MALAYSIA* AKIRA SHIRAI, DAVID M. ROBINSON, GRAHAEM W. BROWN, ELSIE GAN and DAVID L. HUXSOLL U.S. Army Medical Research Unit, Institute for Medical Research, Kuala Lumpur, Malaysia (Received, July 18, Accepted, October 25, 1979) SUMMARY: One hundred and fourteen Rickettsia tsutsugamushi isolates, re covered from febrile patients in central Peninsular Malaysia, were antigenically analyzed by direct immunofluorescence using eight prototype strains. Twenty-nine antigenic types were detected. The TA763, TA716, Karp and TA686 strains were the most common and occurred singly or in combination with each other or other strains in 86% of the isolates. INTRODUCTION Antigenic heterogeneity among strains of Rickettsia tsutsugamushi has been documented by the use of various techniques: (1) complement fixation (Bengtson, 1945, 1946; Shishido, 1964; Shirai and Wisseman, 1975); (2) cross-neutralization (Bell, Bennett and Whitman, 1946; Bennett, Smadel and Gauld, 1947, 1949; Fox, 1949; Miesse, Diercks and Danauskus, 1950; Plotnikova and Tarasevich, 1967); (3) toxin neutralization (Smadel et al., 1946); (4) cross-vaccination (Rights, Smadel and Jackson, 1948); and (5) immunofluorescence (Iida, Kawashima and Kawamura, 1965; Shirai and Wisseman, 1975). The existence of diverse antigenic types within a small geographic area was demonstrated by Miesse et al. (1950), who studied eight strains of R. tsutsugamushi, seven of which were recovered from people who contracted the infection in an 81-hectare field in Malaya and the 8th was from a pool of trombiculid mites collected from a rat in the same field. These workers, using the cross serum neutralization test, placed the eight isolates into four different serogroups. One of many prerequisites in the development of an effective scrub typhus vaccine is the identification of the most prominent antigenic strains associated * Supported by Research Grant No. DAMD17-78-G-9440 from the U.S. Army Medical Research and Development Command, Fort Detrick, Frederick, Maryland 21701, U.S.A. In conducting the research described in this report, the investigators adhered to the gguide for the Care and Use of Laboratory Animals h, as promulgated by the Institute of Laboratory Animal Resources, National Academy of Sciences, National Research Council. 337

2 338 SHIRAI et al. Vol.32 with human disease. Recent investigations of scrub typhus in rural Malaysia (Brown et al., 1976) provided many isolates from febrile patients. This report describes the antigenic analysis of these isolates. MATERIALS AND METHODS Specimens and study sites: From March, 1975 to August, 1976, blood specimens were collected from febrile patients attending either of two district hospitals or a rural health center in central Peninsular Malaysia (Fig. 1). Mentekab district hospital is located in Pahang, while Kuala Pilah district hospital is in Negri Sembilan. The rural health center is located on a Federal Land Development Authority (FELDA) oil-palm plantation at Bukit Mendi, Pahang. Rickettsial isolation: Freshly drawn blood samples were immediately inoculated intraperitoneally into pairs of laboratory mice. Subsequent passages were made, and isolations were confirmed by challenging the mice with the virulent Karp strain of R. tsutsugamushi (Brown et al., 1976). Antigen preparation: For preparation of antigens, mice were initially injected intraperitoneally with 1 ml of a 1% glycogen solution in Hank's BSS. Fig.1. Map of Peninsular Malaysia showing locations of the 3 medical facilities where isolates were obtained.

3 1979 ANALYSIS OF R. TSUTSUGAMUSHI ISOLATES 339 Five days later, they were inoculated intraperitoneally with 0.2 ml of a 20% liver/spleen suspension prepared from infected mice. On day 14 or when the mice became morbid, they were sacrificed and the peritoneal cells were harvested by peritoneal washing with 5 ml of Hanks' BSS. The cells were concentrated by centrifugation at 2,000 ~g for 10 min at 4 C, and the pellet was resuspended in a minimum volume of Hanks' BSS. The pellet suspension was spotted on a clean microscope slide, and dried at room temperature for a minimum of 30 min before fixing with acetone for 10 min. After drying at room temperature, the fixed slides were maintained at -20 C until stained and examined. No slides were stored more than one week before examination. Direct immunoluorescence test: Antigenic analysis of the isolates was done by direct immunofluorescence, as described originally by Iida et al. (1965). The eight strains of R. tsutsugamushi used to prepare antisera for the study have been previously described. The Karp (Derrick and Brown, 1949), Gilliam (Bengtson, 1945) and Kato (Shishido et al., 1958) strains have been accepted as prototype strains (Shishido, 1962, 1964). Studies of many isolates in Thailand have shown the existence of five other distinct strains: TA678, TA686, TA716, TA763 and TH 1817 (Elisberg, Needy and Bozeman, 1978). The fluoresceinconjugated antisera were prepared by inoculating rabbits intraperitoneally with 1 ml of a 20% yolk sac suspension of the respective strains. The serum was collected from the rabbits when the homologous antibody titer, as measured by the indirect immunofluorescent test, reached 1:2,560. The globulin was precipitated with saturated ammonium sulfate (Cherry, Goldman and Carski, 1960). After the globulin was conjugated with fluorescein isothiocyanate, the conjugate was the final dilution at which distinct staining of the rickettsiae could be then absorbed with guinea-pig liver powder before use. Using twofold dilutions, the conjugates were titrated against all eight antigens, and the end-point titer was the final dilution at which distinct staining of the rickettsiae could be observed. Not all conjugated antisera were monospecific in that some reacted with heterologous antigens. In these cases, unlabeled heterologous serum was added to block the heterologous reactivity and insure specificity of the reagent (Elisberg, personal communication). In all instances, the final working dilution of the respective conjugates was two dilutions below the end-point titer. RESULTS The reactions of the conjugates to homologous and heterologous antigens are summarized in Table I. The conjugates of Karp, Kato, TA686, TA716 and TA763 strains were multi-reactive prior to treatment (Pre). In the case of the Karp conjugate, it reacted with Karp, Gilliam, TA716 and TA763 antigens. Therefore, equal amounts of unlabeled, high-titered sera of the three heterologous strains were mixed, and one part of the serum mixture was combined with one part of the Karp conjugate. The reactivity of the conjugate was now monospecific (Post). Gilliam, TA678 and TH1817 conjugates required no treat-

4 340 SHIRAI et al. Vol.32 TABLE I Reactions of 8 prototype strains of Rickettsia tsutsugamushi heterologous antisera in the direct immuno fluorescent test with homologous and 1 Antisera conjugated with fluorescein isothiocyanate. 2 Pre=before addition of unlabeled heterologous antisera.3 P ost-after addition of unlabeled heterologous antisera. 4 Positive fluorescence. 5 Negative fluorescence. ment as they reacted only with the homologous antigens. During the period from March, 1975 to August, 1976, 114 isolates of R. tsutsugamushi were made from blood specimens collected from febrile patients attending the three medical facilities. Fifty-two of 114 isolates were from patients seen at Bukit Mendi health center, while 43 and 19 were those from Mentekab and Kuala Pilah district hospitals, respectively. The isolates were antigenically characterized by the direct fluorescent anti - body test (Table II). Results showed that 89 of 114 (78%) isolates contained mixtures of two or more strains. Tabulating the frequency in which each prototype strain of R. tsutsugamushi occurs among the isolates demonstrated that the predominant strains were TA763 and TA716 (Table III). The antigenic make-up of the isolates from the three medical treatment facilities was similar. DISCUSSION The findings of this study substantiated the existence of antigenic hetero - geneity of scrub typhus isolates recovered from febrile patients in central Peninsular Malaysia. All 114 isolates reacted with at least one of the eight strains, and 89 (78%) reacted with two or more strains. On the basis of these reactions, 29 antigenic types were detected (Table II). We do not know whether

5 1979 ANALYSIS OF R. TSUTSUGAMUSHI ISOLATES 341 TABLE II Antigenic characterization o f human isolates from Bukit Mendi HealthCenter, Mentekab District Hospital and Kuala Pilah District Hospital 1 Number of isolates (%). the antigenic combinations which reacted with two or more prototype strains comprise a single organism with a mosaic of antigenic determinants or a mixture of organisms. There is evidence to suggest that recently isolated strains may contain mixtures. Elisberg et al. (1968) found that antigenic determinants were lost from human isolates upon serial passage in laboratory mice. The development of a plaque purification technique (Oaks et al., 1979) will now enable investigators to address the problem of multiple antigenicity. Whether or not our human isolates were mixtures of rickettsial strains, the occurrence of the Gilliam strain was limited. Of the 25 isolates in which only

6 342 SHIRAI et al. Vol.32 TABLE III Frequency of Rickettsia tsutsugamushi prototype antigens found among the human isolates 1 Expressed as the percentage of isolates in which the respective antigen was identified. one antigen was detected, 14 reacted with Gilliam. Furthermore, the Gilliam antigen appeared in combination with other antigens in only six isolates. Thus, Gilliam was not a common antigen and tended to exist alone when it appeared. The TA763, TA716, Karp and TA686 strains were the most common and occurred singly or in combination with each other or other strains in 86% of the isolates. Elisberg et al. (1978) recently reported the antigenic interrelationships among these eight strains of R. tsutsugamushi by indirect immunofluorescence. They observed that Karp, Kato, TA686 TA716, TA763 and TH 1817 strains shared disproportionate but significant antigens. The most significant amounts, however, were shared among TA686, TA716 and TA763 strains. Additionally, antibody responses in monkeys experimentally infected with Karp, TA686, TA716 or TA763 strain have provided evidence that these strains are antigenically related (unpublished data). Further demonstration of antigenic similarity was shown in the reactions of untreated antisera to Karp, TA686, TA716 and TA763 strains (Table I), in which heterologous reactions occurred among the four strains. Shirai and wisseman (1975) studied 79 R. tsutsugamushi isolates from Pakistan by using complement fixation and direct immunofluorescence techniques for comparison with the Karp, Gilliam and Kato strains. They found that the majority of the isolates reacted with the Karp strain, although in many instances the degree of reactivity was low. One can speculate that many of the isolates, which showed low reactivity to Karp, may have been one of the Karp-related strains. Antigenic heterogeneity has contributed to difficulties in preparing an effective scrub typhus vaccine. In fact, the identification of multiple antigenic types, or sero-groups, has often discouraged the pursuit of a vaccine. With few exceptions, most previous studies demonstrating antigenic heterogeneity among

7 1979 ANALYSIS OF R. TSUTSUGAMUSHI ISOLATES 343 R.tsutsugamushi strains were done using limited numbers of isolates and often did not include a comparison with established prototype strains. In the current study, 114 human isolates were examined. All reacted with at least one of the prototype strains; however, none reacted with TA6 78 and only a few reacted with the Kato strain. Thus, of the 29 antigenic types identified, only six antigens, existing singly or in combination, accounted for 27 different antigenic types. If the majority of the infections throughout the endemic region are caused by a few related antigens, difficulties in selecting strains to be incorporated into vaccines may be overcome. ACKNOWLEDGEMENTS The authors would like to thank Dr. George F. de Witt, Director, Institute for Medical Research, for his enthusiastic support and his permission to publish this paper, and the technical staff of USAMRU for their valuable assistance. REFERENCES BELL, E. J., BENNETT, B. L. AND WHITMAN, L. (1946): Antigenic differences between strains of scrub typhus as demonstrated by cross-neutralization tests. Proc. Soc. Exptl. Biol. Med., 62, BENGTSON, I. A. (1945): Apparent serological heterogeneity among strains of tsutsugamushi disease (scrub typhus). Public Health Rep., 60, BENGTSON, I. A. (1946): A serological study of 37 cases of tsutsugamushi disease (scrub typhus) occurring in Burma and the Philippine Islands. Public Health Rep., 61, BENNETT, B. L., SMADEL, J. E. AND GAULD, R. L. (1947): Differences in strains of Rickettsia orien talis as demonstrated by cross-neutralization tests. J. Bacterial., 54, 93. BENNETT, B. L., SMADEL, J. E. AND GAULD, R. L. (1949): Studies on scrub typhus (tsutsugamushi disease). IV. Heterogeneity of strains of R. tsutsugamushi as demonstrated by cross-neutralization tests. J. Immunol., 62, BROWN, G. W., ROBINSON, D. M., HUXSOLL, D. L., NG, T. S., LIMA K. J. AND SANNASEY, G. (1976): Scrub typhus: a common cause of illness in indigenous populations. Trans. Roy. Soc. Trop. Med. Hyg., 70, CHERRY, W. B., GOLDMAN, M. AND CARSKI, T. R. (1960): Fluorescent antibody techniques in the diagnosis of communicable diseases. U.S. Public Health Serv. Publ., 729, DERRICK, E. H. AND BROWN, H. E. (1949): Isolation of the Karp strain of Rickettsia tsutsuga mushi. Lancet, 2, ELISBERG, B. L., CAMPBELL, J. M. AND BOZEMAN, F. M. (1968): Antigenic diversity of Rickettsia tsutsugamushi: epidemiologic and ecologic significance. J. Hyg. Epid. Microbiol. Immunol., 12, ELISBERG, B. L., NEEDY, C. F. AND BOZEMAN, F. M. (1978): Antigenic interrelationships among strains of Rickettsia tsutsugamushi. p In J. KAZAR, R. A. ORMSBEE and I. N. TARASE VICH [eds.], Rickettsiae and Rickettsial Diseases. VEDA Publ. House of the Slovak Acad. Sci., Bratislava.. Fox, J. P. (1949): The neutralization technique in tsutsugamushi disease (scrub typhus) and the antigenic differentiation of rickettsial strains. J. Immunol., 62, IIDA, T., KAWASHIMA, H. AND KAWAMURA, A. (1965): Direct immunofluorescence for typing of tsutsugamushi disease rickettsia. J. Immunol., 95, MIESSE, M., DIERCKS, F. H. AND DANAUSKAS, J. X. (1950): Strain differences among Rickettsia tsutsugamushi (abstract). Bact. Proc., OAKS, S. C., JR., OSTERMAN, J. V. AND HETRICK, F. M. (1977): Plaque assay and cloning of scrub typhus rickettsiae in irradiated L-929 cells. J. Clin. Microbiol., 6,

8 344 SHIRAI et al. Vol.32 PLOTNIKOVA, L. F. AND TARASEVICH, I. V. (1967): Antigennia struktura nekotorykh shtammov rikketsii tsutsugamushi (v opytakh perekrestoni reaktsii neitralizatii). Zh. Mikrobiol. Epidemiol. Immunobiol., 9, RIGHTS, F. L., SMADEL, J. E. AND JACKSON, E. B. (1948): Studies on scrub typhus (tsutsugamushi disease). III. Heterogenicity of strains of R. tsutsugamushi as demonstrated by cross-vaccination studies. J. Exptl. Med., 87, SHIRAI, A. AND WISSEMAN, C. L. JR. (1975): Serologic classification of scrub typhus isolates from Pakistan. Am. J. Trop. Med. Hyg., 24, SHISHIDO, A., OHTAWARA, M., TATENO, S., MIZUNO, S., OGURA, M. AND KITAOKA, M. (1958): The nature of immunity against scrub typhus in mice. I. The resistance of mice, surviving sub cutaneous infection of scrub typhus rickettsia, to intraperitoneal reinfection of the same agent. Japan. J. Med. Sci. Biol., 11, SHISHIDO, A. (1962): Identification and serological classification of the causative agent of scrub typhus in Japan. Japan. J. Med. Sci. Biol., 15, SHISHIDO, A. (1964): Strain variation of Rickettsia orientalis in the complement fixation test. Japan. J. Med. Sci. Biol., 17, SMADEL, J. E., JACKSON, E. B., BENNETT, B. L. AND RIGHTS, F. L. (1946): A toxic substance asso ciated with the Gilliam strain of R. orientalis. Proc. Soc. Exptl. Biol. Med. 62,