- Email: [email protected]
Molecular comparison of antigens and proteins of virulent and avirulent clones of leptospira interrogans serovar copenhageni, strain Shibaura
Zbl. Bakt. Hyg. A 266,453-462 (1987)
Molecular Comparison of Antigens and Proteins of Virulent and Avirulent Clones of Leptospira interrogans Serovar copenhageni, Strain Shibaura
MASAHIRO NIIKURA, ETSURO ONO, and RYO YANAGAWA Department of Hygiene and Microbiology, Faculty of Veterinary Medicine, Hokkaido University, Sapporo 060, Japan (Head: Prof. Or. R. Yanagawa)
With 4 Figures· Received April 15, 1986 . Accepted April 13, 1987
Abstract The antigens and the protein profiles of virulent and avirulent clones of Leptospira interrogans serovar copenhageni, strain Shibaura were compared by SOS-polyacrylamide gel electrophoresis (SOS-PAGE), Western blotting and radioimmunoprecipitation. An antigen specific to the virulent clone was detectable only in the virulent clone by Western blotting using anti-virulent clone but not anti-avirulent clone antisera. The molecular weight of the antigen was estimated to be approximately 32000 by SOS-PAGE,and the antigen was found to be thermostable. A 43000 dalton antigenic molecule was found to be specific for the virulent clone, and the 41000 dalton antigenic molecule was found to be specific for the avirulent clone by Western blotting using both anti-virulent and anti-avirulent clones antisera. Therefore, these antigenic molecules seemed to contain at least one antigenic determinant which is common to the virulent and the avirulent clones. No difference between the virulent and avirulent clones was observed in the comparison of protein profiles on SOS-PAGEand surface exposed antigenic protein profiles on radioimmunoprecipitation using radioiodinated leptospiral surface exposed proteins. Therefore, the antigens which were different between the virulent and avirulent clones were suspected not to be protein antigens.
Zusammenfassung Die Antigene und Proteinprofile virulenter und avirulenter Klone von Leptospira interrogans Serovar copenhageni, Stamm Shibaura wurden mit Hilfe von SOS-PolyacrylamidgelElektrophorese (SOS-PAGE), Western Blot und Radioimmunprazipitation verglichen. Ein fiir den virulenten Klon spezifisches Antigen wurde nur im virulenten Klon beim Western Blot-Verfahren unter Verwendung von Antiseren gegen den virulenten Klon, aber nicht von Antiseren gegen den avirulenten Klon festgestellt. Das Molekulargewicht wurde mittels SOS-PAGEauf ca. 32000 geschatzt; das Antigen war thermostabil. Ein Antigenmolekiil von 43000 Dalton war fiir den virulenten Klon spezifisch und eines von 41 000 Dalton fiir den avirulenten Klon, wenn das Western Blot-Verfahren unter Verwendung von Antiseren gegen den virulenten und den avirulenten Klon angewandt wurde. Somit schienen diese Antigen-
454
M. Niikura, E.Ono, and R. Yanagawa
molekiile mindestens eine bei den virulenten und den avirulenten Klonen gleichzeitig vorhandene Antigendeterminante zu enthalten. Beim Vergleich der Proteinprofile bei SDSPAGE und der oberfliichenexponierten Antigen-Protein-Profile bei der Radioimmunpriizipitation unter Verwendung Radiojod-markierter oberfliichenexponierter Leptospiren-Proteine wurde kein Unterschied zwischen virulenten und avirulenten Klonen beobachtet. Es wird daher vermutet, dag es sich bei den bei virulenten und avirulenten Klonen verschiedenen Antigenen nicht urn Protein-Antigene handelte.
Introduction
There has been only one report on the antigenic profiles of Leptospira. A study of proteins of several strains of serovar hardjo (10) revealed that most of the major protein antigens were present in the outer envelope, and a number of major protein antigens were cross-reactive with strains of serovars balcanica and pomona. No information, however, is available on the antigenic protein profiles which are different between virulent and avirulent clones of a strain. There have been reports on a comparison of the colonial morphology (17) and antigenicity (14) of virulent and avirulent clones of leptospira strains. A thermolabile antigen has been detected by the agglutinin-absorption test and the precipitin-absorption test in gel only in the virulent clone of 1. interrogans serovar copenhageni, strain Shibaura and a virulent strain of serovar pomona (14). The presence of a Vi antigen of leptospira, which is also thermolabile, has been reported (5), but the antigen has not been isolated. The present study describes a molecular comparison of the antigens and proteins of virulent and avirulent clones of 1. interrogans, and serovar copenhageni, strain Shibaura, by SDS-polyacrylamide gel electrophoresis (SDS-PAGE), Western blotting and radioimmunoprecipitation.
Materials and Methods Strains. The strains used were virulent and avirulent clones of L. interrogans serovar copenhageni, strain Shibaura (13, 17). The virulent clone was passaged in guinea pigs every 2 months and shown to be lethal for guinea pigs. The avirulent clone, derived from the virulent clone, was passaged in media for more than 15 years and determined to be avirulent. These strains were grown at 30°C in 0.2 % tryptose phosphate broth (Difco, Detroit, MI) containing 10% normal rabbit serum. Immune sera. Hyperimmune rabbit serum against the virulent and avirulent clones was prepared as previously described (16). The microscopic agglutination titres of the antivirulent clone antiserum and anti-avirulent clone antiserum were 1:102400 and 1:204800, respectively. Leptospiral extract for SDS-polyacrylamide gel electrophoresis (SDS-PA GE) and Western blotting. Leptospiral extracts were prepared as described by Yanagawa et al. (16) with a slight modification. Briefly, 4 X 1011 organisms were extracted with 1 ml of 0.5% SDS at 45°C for 4 h under shaking. Then, the extract was centrifuged at 15000 rpm on a Microfuge (Tomy Seiko, Tokyo) for 4 h, and the supernatant was used as the leptospiral extract sample. Absorption of anti-virulent clone antiserum with the avirulent clone. The anti-virulent clone antiserum was absorbed with the avirulent clone leptospiras as follows. 1 ml of avirulent clone leptospiras (1.0 X 1010 organisms/ml) suspended in phosphate-buffered
Leptospiral Antigens
455
saline (PBS, pH 7.4) was added to 25 Itl of anti-virulent clone antiserum, whose microscopic agglutination titre was 1:102400 in relation to the virulent clone. The mixture was incufor 1 h with occasional shaking and thereafter centrifuged. Further absorpbated at tion of the supernatant was carried out with 0.5 ml of the antigen. The absorption procedure was repeated until the supernatant did not agglutinate the avirulent clone at the dilution of 1:4, because of the high sensitivity of Western blotting. Ten ml of the leptospiral suspension was needed to absorb the anti-virulent clone antiserum to this level. Therefore, if we considered the dilution factor (25 1t11l0 ml = 400) alone, the titre of the original antivirulent clone antiserum to the virulent clone should be theoretically calculated as 1:256 (102400/400). Actually, after the absorption procedure, the titre of the antiserum to the virulent clone was 1:32. The reduction of the titre of the absorbed antiserum may have been due to factors of absorption. The same absorbed antiserum did not agglutinate the avirulent clone at a dilution of 1:4. The anti-avirulent clone antiserum which was absorbed with the virulent clone leptospiras showed no difference in the microscopic agglutination titres between the virulent clone and the avirulent clone. SDS-polyacrylamide gel electrophoresis (SDS-PAGE). Leptospiral extracts were separated on 12% polyacrylamide slab gels in the discontinuous Tris-glycine system as described by Laemmli (6). The samples were not boiled before electrophoresis. Gels were fixed and stained with Coomassie brilliant blue. The molecular weights were estimated by the method of Weber and Osborn (15). Western blotting. Western blotting was performed by the technique of Towbin et al. (12) with a Trans Blot Cell (Bio-Rad, Richmond, Ca.). Proteins were transferred to nitrocellulose membrane (Bio-Rad, Richmond, Ca.) at 15 rnA for 1 h following 30 rnA for 12 h with cooling in 25 mM Tris-192 mM glycine-12% (v/v) methanol buffer. The sheets were incubated with PBS (pH 7.4) containing 3% of bovine serum albumin (BSA, Boehringer Mannheim, Mannheim) for 1 h at room temperature to reduce nonspecific reaction. The sheets were incubated with anti-virulent clone or anti-avirulent clone antiserum at a dilution of 1/ 100 (or anti-virulent clone antiserum absorbed with avirulent clone at a dilution of 1/10) in 0.5% BSA-PBS for 2 h with constant rocking. Then, the sheets were rinsed with 0.05% Tween 20-PBS (PBST) and incubated with horseradish-peroxidase-conjugated anti-rabbit IgG (heavy and light chains, Bio-Rad, Richmond, Ca.) in 0.5% BSA-PBS at a dilution of 1/ 1000 for 2 h with constant rocking. The sheets were rinsed with PBSTand incubated with 4chloro-l-naphtol in 10 mM Tris-HCI buffer containing 100 mM NaCI (pH 8.0) for 10 to 20 min. Radiolabelling of leptospiras. Lactoperoxidase-catalyzed iodination of intact leptospiras was performed as described (1). Briefly, 0.5 mCi of Na 1251 (Amersham, Backinghamshire) and 100 Itl of lactoperoxidase (Sigma, St. Louis, MO.) were gently mixed with 0.5 ml of PBS containing 5.0 X 10 10 organisms. Then, 50 ul of H 202 solution (5 [!l of 30% stock H 202 freshly prepared in 125 ml of PBS) were induced, and the mixture was incubated for 15 min. Another 50 ul of H 2 0 2 solution were added and the reaction was allowed to stand for 15 min. The iodinated organisms were then washed and extracted with 0.5 ml of 0.5% SDS at 45 DC for 1 h. The extract was centrifuged at 15000 rpm on a Microfuge for 30 min, and the supernatant was used as radioiodinated leptospiral extract sample. Immunoprecipitation of radio iodinated leptospiral proteins. Immunoprecipitation was carried out by a modification of the procedure of Nunes-Edwards et al. (10). Biefly, 50 ul of the radioiodinated leptospiral extract were mixed with 450 [!lof Triton buffer (11) and then 30 ul of normal or test antisera were added. When the absorbed serum was used as the test serum, 90 Itl were added because the absorbed serum was rather low in its agglutination titre. After 1 h incubation at room temperature, 75 Itl of a slurry of protein A-Sepharose CL4B (Pharmacia, Uppsala) were added. The mixture was incubated at room temperature for 1 h with gentle shaking, pelleted and washed sufficiently with Triton buffer and then with 10 mM Tris-HCI buffer (pH 8.0). The immunoprecipitate was extracted from the pellet by adding 50 ul of SDS sample buffer (6) without 2-mercaptoethanol, heated in boiling water
3rc
456
M. Niikura, E. Ono, and R. Yanagawa
for 3 min and centrifuged in a Microfuge. A supernatant containing the precipitate was collected and analyzed by SDS-PAGE as previously described.
Results Comparison of protein profiles of virulent and avirulent clones by SDS-PAGE Protein profiles of virulent and avirulent clones of 1. interrogans serovar copenhageni, strain Shibaura were compared by SDS-PAGE (Fig. 1). More than 30 protein bands were detectable on either virulent or avirulent clones. The protein profiles of the virulent and avirulent clones were identical. Comparison of antigen profiles of virulent and avirulent clones by Western blotting 1) Comparison with anti-virulent clone antiserum. Anti-virulent clone antiserum was used first to identify the antigens of the virulent and avirulent clones (Fig. 2). More than 30 bands were visualized on the extracts from the virulent and avirulent clones,
MW _ 9 2 .5 K
-66 .2K
_45K
_31K
a
b
Fig. 1. Comparison of the protein profiles of the virulent (lane a) and avirulent (lane b) clones of 1. interrogans serovar copenhageni strain Shibaura. Samples were extracted with 0.5% SDS at 45°C for 4 hours and separated on 12% SDS-polyacrylamide gel. The gel was stained with Coomassie brilliant blue. No obvious differences in the position of bands are evident. The molecular weight markers appearing on the right represent the following protein standards: phosphorylase B, 92500; bovine serum albumin, 66200; ovalbumin, 45000; and carbonic anhydrase, 31000.
Leptospiral Antigens
457
MW
_92 .5K
_66.2K
a
b
c
d
Fig. 2. Comparison of antigen profiles of the virulent and avirulent clones by Western blotting using anti-virulent clone antiserum. Samples were extracted with 0.5% SDS at 45°C for 4 hours and separated on 12% SDS-polyacrylamide gel. Western blotting was performed as described in the text. Lanes: a. Virulent clone of L. interrogans serovar copenbageni strain Shibaura; b. Avirulent clone of L. interrogans serovar copenbageni strain Shibaura; c. Virulent clone of L. interrogans serovar copenhageni strain Shibaura heated at 100°C for 10 minutes before electrophoresis; d. Avirulent clone of L. interrogans serovar copenhageni strain Shibaura heated at 100°C for 10 minutes before electrophoresis. The bands which were different for the virulent and avirulent clones are indicated on the left. Molecular weight markers are given in the legend to Fig. 1.
respectively. Most of the antigen bands were identical, but there were two clear differences: 1) An antigen of about 32000 dalton was detected on the extract of the virulent clone but not on the extract of the avirulent clone; and 2) A 43000 dalton antigen was strongly stained only on the extract of the virulent clone, while the 41000 dalton antigen was strongly stained only on the avirulent clone. The 32000, 41000 and 43000 dalton antigens were not detected on the protein profiles on SDS-PAGE. 2) Comparison with anti-avirulent clone antiserum. Anti-avirulent clone antiserum was used to identify the antigens of the virulent and avirulent clones (Fig. 3). More than 30 bands were visualized on the extracts from the virulent and avirulent clones, respectively, by anti-avirulent clone antiserum as well as by anti-virulent clone antiserum. The 43000 dalton antigen on the virulent clone extract and the 41000 dalton
458
M. Niikura, E.Ono, and R. Yanagawa MW
_66.2K
--
_45K
_31K
a
b
c
d
Fig. 3 Comparison of antigen profiles of the virulent and avirulent clones by Western blotting using anti-avirulent clone antiserum. Samples was extracted with 0.5% SDS at 45°C for 4 hours and separated on 12% SDS-polyacrylamide gel. Western blotting was performed as described in the text. Lanes: a. Virulent clone of L. interrogans serovar copenhageni strain Shibaura; b. Avirulent clone of L. interrogans servar copenhageni strain Shibaura; c. Virulent clone of L. interrogans serovar copenhageni strain Shibaura heated at 100°C for 10 minutes before electrophoresis; d. Avirulent clone of L. interrogans serovar copenhageni strain Shibaura heated at 100°C for 10 minutes before electrophoresis. The bands which were different for the virulent and avirulent clones are indicated on the left. Molecular weight markers are given in the legend to Fig. 1. antigen on the avirulent clone extract were also detected with anti-avirulent clone antiserum. The 32000 dalton antigen found in the virulent clone extract by utilizing anti-virulent clone antiserum was not detected by utilizing the anti-avirulent clone antiserum. Heating the extract at 100°C for 10 min did not affect the antigen profiles of the extracts identified by both anti-virulent and avirulent clones antisera (Figs. 2 and 3, lanes c and d). Only a slight band of 45000 dalton was detected in the extracts of both clones with anti-virulent clone antiserum absorbed with the avirulent clone. The 32000 dalton antigen was not detected. No antigens were detected in the Western blotting experiments performed with normal rabbit serum as control. All the experiments using the Western blotting method were repeated more than three times and the same results were obtained.
Leptospiral Antigens
459
Comparison of surface-exposed antigenic protein by radioimmunoprecipitation using 125/ labelled leptospira 125
1 labelled organisms were examined by radioimmunoprecipitation to determine whether or not the 32000, 41000 and 43000 dalton antigens were surface-exposed proteins. As shown in Fig. 4, 16 proteins on the virulent clones extracts were radioiodinated (Fig. 4, lanes a and b). Both antisera precipitated 13 proteins of 93000, 76000, 66000, 62000, 58000, 46000, 40500, 36000, 33000, 29000, 23000, 18000 and MW
_
92 .5 K
_
66. 2 K
_45K
a
bed
e
f
9
h
Fig. 4. Comparison of surface-exposed antigenic proteins of the virulent and avirulent clones by radioimmunoprecipitation. Samples were SDS-soubilized 1251 labelled virulent (lanes a, c, e, g and i) and avirulent (b, d, f, hand j) clones of L. interrogans serovar copenhageni strain Shibaura. Lanes: a. 125I-labelled protein profile of the virulent clone; b. 125I-labelled protein profile of the avirulent clone; c. 12sI-labelled antigenic protein profile of the virulent clone precipitated with anti-virulent clone antiserum; d. 125I-labelled antigenic r:rotein profile of the avirulent clone precipitated with the anti-virulent clone antiserum; e. 2sI-labelled antigenic protein profile of the virulent clone precipitated with the anti-avirulent clone antiserum; f. 12sI-labelled antigenic protein profile of the avirulent clone precipitated with the anti-avirulent clone antiserum; g. 125I-labelled protein profile of the virulent clone precipitated with the anti-virulent clone antiserum absorbed with the avirulent clone; h. 12sl-labelled protein profile of the avirulent clone precipitated with anti-virulent clone antiserum absorbed with the avirulent clone; i. WI-labelled protein profile of the virulent clone precipitated with normal rabbit serum; j. WI-labelled protein profile of the avirulent clone precipitated with normal rabbit serum. No obvious differences are evident among lanes a, b, c and d. Molecular weight markers are given in the legend to Fig. 1.
460
M. Niikura, E. Ono, and R. Yanagawa
16000 dalton on the extracts of both virulent and avirulent clones (Fig. 4, lanes c and d; e and f). Of these proteins, however, 2 proteins of 93000 and 16000 dalton were precipitated with normal rabbit serum (Fig. 4, lanes i and i), indicating that these 2 proteins were precipitated non-specifically. The 32000, 41000 and 43000 dalton antigens were not precipitated with the anti-virulent and anti-avirulent clones antisera, suggesting that these antigens were not surface-exposed proteins. Anti-virulent clone antiserum absorbed with the avirulent clone precipitated 2 proteins, which were identical to those precipitated with normal rabbit serum, showing that the 2 proteins were non-specifically precipitated (Fig. 4, lanes g and h). This fact may indicate that the so called "Vi antigen", which has been thought to be present on leptospiral cells of the virulent clone and was detectable by the microscopic agglutination test, was not demonstrated by radioimmunoprecipitation.
Discussion The antigen specific to the virulent clone of L. interrogans serovar copenhageni, strain Shibaura was detected in the present study by Western blotting. The 32000 dalton antigen specific to the virulent clone was thought to be of non-protein nature because no corresponding band was detected on the protein profiles on SDS-PAGE stained with Coomassie brilliant blue. This antigen was considered not to be a surface antigen protein because no corresponding band was detected by radioimmunoprecipitation, A lipopolysaccharide in Haemophilus influenzae is a precedential example of a non-protein antigen, which was detected by Western blotting (7). The chemical nature of the 32000 dalton antigen will be a subject of future study. The 32000 dalton antigen was not detected, contrary to expectation, by Western blotting with the absorbed serum. One explanation may be as follows. The 32000 dalton antigen found only in the virulent clone (but not large in amount) may be present in the avirulent clone in such a small amount as to be undetectable by Western blotting and the antibodies to the antigen may have been absorbed during the absorption procedure. The 43000 dalton antigen molecule was detected only in the virulent clone, while the 41000 dalton antigen molecule was detected only in the avirulent clone. These antigenic molecules were, however, detected with both anti-virulent and anti-avirulent clone antisera. These antigenic molecules of the virulent and avirulent clones may share at least one common antigenic determinant, but they migrated to different positions on SDS-PAGE. Antigens which have a several common antigenic determinant(s) and different molecular weight were reported to exist in strains of Haemophilus influenzae (4). The 43000 and 41000 dalton antigen molecules detected by Western blotting, however, had no corresponding bands on either SDS-PAGE stained with Coomassie brilliant blue or radioimmunoprecipitation using 1251 labelling. These antigen molecules were therefore considered to be non-protein antigens. The 32000 dalton antigen specific to the virulent clone was found to be thermostable. The 32000 dalton antigen may, therefore, be different from the "Vi antigen". Ueno et al. (14) reported that the thermolabile "Vi antigen" was found in the virulent clone of L. tnterrogans serovar copenhageni, strain Shibaura by the microscopic agglutination test with the anti-virulent clone antiserum absorbed with the avirulent clone organisms. The agglutination titre of "Vi antigen" detected by the absorbed serum was rather low. In the present study, "Vi antigen" was not detected by Western
Leptospiral Antigens
461
blotting, but detected in the microscopic agglutination test. The fact that the thermolabile antigen was detected by the microscopic agglutination test agreed with a number of reports showing that leptospiras have thermolabile antigens (2, 8, 9). The reasons why "Vi antigen" was not detected by Western blotting is not clear. However, some possible reasons why "Vi antigen" was not detected by Western blotting could be: 1) Thermolabile and thermostable antigenic determinants may coexist on an antigen molecule; 2) Thermolabile and thermostable antigens may have comigrated to the same position in SDS-PAGE; and 3) Several antigenic molecules including thermolabile antigen may not have been transferred to the nitrocellulose sheet. Differences in the colony type, susceptibility to leptospiricidal activity mediated by antibody plus complement (17) and ability to attach to cultured cells (13) were reported to exist between the virulent and avirulent clones. How the 32000, 41000 and 43000 dalton antigens, which were different in the virulent and avirulent clones, correlate to such properties and to virulence is another subject for future study. Acknowledgments. This work was supported by Grants-in-Aid for Scientific Research Nos. 60440020 and 60760221 from the Ministry of Eduation, Science and Culture of Japan. All experiments utilizing radioisotopes could be carried out at the Central Institute of Radioisotope Science, Hokkaido University, through the cooperation kindly provided by Prof. S. Nishi and Dr. T. Ohnishi.
References
1. Alderete, ]. F. and]. B. Baseman: Surface characterization of virulent Treponema pallidum. Infect. Immun. 30 (1980) 814-823 2. Borg-Petersen, c.: A thermo-labile antigen in the leptospira strain Ictero No. 1. Trop. geogr. Med. 23 (1971) 282-285 3. Paine, S.: Leptospirosis. 37-98. In: Guidelines for the control of leptospirosis. Ed. S. Faine World Health Organization, Geneva (1982) 4. Gulig, P. A., G. H. McCracken jr., c. F. Frisch, K. H. Johnson, and E. ]. Hansen: Antibody response of infants to cell surface exposed outer membrane protein of Haemophilus influenzae Type B after systemic Haemophilus disease. Infect. Immun. 37 (1982) 82-88 5. Kmety, E.: Ober ein Vi-Antigen bei Leptospiren. Zbl. Bakt. Hyg., I. Abt, Orig. A 211 (1972) 343-351 6. Laemmli, U. K.: Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227 (1970) 680-685 7. Loeb, M. R.: Immunoblot method for identifying components, determinating their cross-reactivity, and investigating cell topology: Results with Haemophilus influenzae Type B. Analyt, Biochem. 143 (1984) 196-204 8. Manev, c.: Serological characterization of the leptospira serogroup pomona. I. Factor analysis of the reference strains. Zbl. Bakt. Hyg., I. Abt. Orig. A 236 (1976) 316-322 9. Nicolescu, M. and G H. Moldoveanu: The thermolabile antigen of some reference leptospira strains from the pomona serogroup. Arch. Roum. Path. Exp. Microbiol. 36 (1977) 151-155 10. Nunes-Edwards, P. L., A. B. Thiermann, P.]. Bassford [r., and L. V. Stamm: Identification and characterization of the protein antigen of Leptospira interrogans serovar hard;0. Infect. 1mmun. 48 (1985) 482-497
462
M. Niikura, E.Ono, and R. Yanagawa
11. Stamm, L. V., T. C. Kerner jr., V. A. Bankaitis, and P. J. Bassford jr.: Identification and preliminary characterization of Treponema pallidum protein antigens expressed in Escherichia coli. Infect. Immun. 41 (1983) 709-721 12. Toiobin, H., T. Staehelin, and J. Gordon: Electrophoretic transfer of protein from polyacrylamide gel to nitrocellulose sheets: Procedure and some applications. Proc. Nat. Acad. Sci. USA 76 (1979) 4350-4354 13. Tsuchimoto, M., M. Niikura, E. Ono, and R. Yanagawa: Leptospiral attachment to cultured cells. Zbl. Bakt. Hyg., I. Abt. Orig. A 258 (1984) 268-274 14. Ueno, K., R. Yanagaioa, and H. Kida: Presence of Vi antigen in a virulent strain of Leptospira interrogans serovar pomona and relation of Vi antigens of leptospira to resistance to leptospiricidal activity mediated by antiserum plus complement. Zbl. Bakt. Hyg., I. Abt, Orig. A 252 (1982) 557-565 15. Weber, K. and M. Osborn: The reliability of molecular weight determinations by dodecyl sulfate-polyacrylamide gel electrophoresis. J. BioI. Chern. 244 (1964) 4406-4412 16. Yanagauia, R., M. Shinagawa, and I. Takashima: Serological studies of leptospires by immunodiffusion. Zbl. Bakt. Hyg., I. Abt. Orig. A 228 (1974) 369-377 17. Yepez-Plascencia, W., R. Yanagauia, and K. Ueno: Correlation of virulence, susceptibility to leptospiricidal activity test mediated by antiserum plus complement and colonial morphology of five lines of a strain of Leptospira interrogans serovar copenhageni. Zbl. Bakt. Hyg., I. Abt, Orig. A 251 (1981) 230-256 Professor Dr. Ryo Yanagatoa, Hokkaido University, Faculty of Veterinary Medicine, Dept. of Hygiene and Microbiology, Sapporo 060, Japan
Molecular Comparison of Antigens and Proteins of Virulent and Avirulent Clones of Leptospira interrogans Serovar copenhageni, Strain Shibaura
MASAHIRO NIIKURA, ETSURO ONO, and RYO YANAGAWA Department of Hygiene and Microbiology, Faculty of Veterinary Medicine, Hokkaido University, Sapporo 060, Japan (Head: Prof. Or. R. Yanagawa)
With 4 Figures· Received April 15, 1986 . Accepted April 13, 1987
Abstract The antigens and the protein profiles of virulent and avirulent clones of Leptospira interrogans serovar copenhageni, strain Shibaura were compared by SOS-polyacrylamide gel electrophoresis (SOS-PAGE), Western blotting and radioimmunoprecipitation. An antigen specific to the virulent clone was detectable only in the virulent clone by Western blotting using anti-virulent clone but not anti-avirulent clone antisera. The molecular weight of the antigen was estimated to be approximately 32000 by SOS-PAGE,and the antigen was found to be thermostable. A 43000 dalton antigenic molecule was found to be specific for the virulent clone, and the 41000 dalton antigenic molecule was found to be specific for the avirulent clone by Western blotting using both anti-virulent and anti-avirulent clones antisera. Therefore, these antigenic molecules seemed to contain at least one antigenic determinant which is common to the virulent and the avirulent clones. No difference between the virulent and avirulent clones was observed in the comparison of protein profiles on SOS-PAGEand surface exposed antigenic protein profiles on radioimmunoprecipitation using radioiodinated leptospiral surface exposed proteins. Therefore, the antigens which were different between the virulent and avirulent clones were suspected not to be protein antigens.
Zusammenfassung Die Antigene und Proteinprofile virulenter und avirulenter Klone von Leptospira interrogans Serovar copenhageni, Stamm Shibaura wurden mit Hilfe von SOS-PolyacrylamidgelElektrophorese (SOS-PAGE), Western Blot und Radioimmunprazipitation verglichen. Ein fiir den virulenten Klon spezifisches Antigen wurde nur im virulenten Klon beim Western Blot-Verfahren unter Verwendung von Antiseren gegen den virulenten Klon, aber nicht von Antiseren gegen den avirulenten Klon festgestellt. Das Molekulargewicht wurde mittels SOS-PAGEauf ca. 32000 geschatzt; das Antigen war thermostabil. Ein Antigenmolekiil von 43000 Dalton war fiir den virulenten Klon spezifisch und eines von 41 000 Dalton fiir den avirulenten Klon, wenn das Western Blot-Verfahren unter Verwendung von Antiseren gegen den virulenten und den avirulenten Klon angewandt wurde. Somit schienen diese Antigen-
454
M. Niikura, E.Ono, and R. Yanagawa
molekiile mindestens eine bei den virulenten und den avirulenten Klonen gleichzeitig vorhandene Antigendeterminante zu enthalten. Beim Vergleich der Proteinprofile bei SDSPAGE und der oberfliichenexponierten Antigen-Protein-Profile bei der Radioimmunpriizipitation unter Verwendung Radiojod-markierter oberfliichenexponierter Leptospiren-Proteine wurde kein Unterschied zwischen virulenten und avirulenten Klonen beobachtet. Es wird daher vermutet, dag es sich bei den bei virulenten und avirulenten Klonen verschiedenen Antigenen nicht urn Protein-Antigene handelte.
Introduction
There has been only one report on the antigenic profiles of Leptospira. A study of proteins of several strains of serovar hardjo (10) revealed that most of the major protein antigens were present in the outer envelope, and a number of major protein antigens were cross-reactive with strains of serovars balcanica and pomona. No information, however, is available on the antigenic protein profiles which are different between virulent and avirulent clones of a strain. There have been reports on a comparison of the colonial morphology (17) and antigenicity (14) of virulent and avirulent clones of leptospira strains. A thermolabile antigen has been detected by the agglutinin-absorption test and the precipitin-absorption test in gel only in the virulent clone of 1. interrogans serovar copenhageni, strain Shibaura and a virulent strain of serovar pomona (14). The presence of a Vi antigen of leptospira, which is also thermolabile, has been reported (5), but the antigen has not been isolated. The present study describes a molecular comparison of the antigens and proteins of virulent and avirulent clones of 1. interrogans, and serovar copenhageni, strain Shibaura, by SDS-polyacrylamide gel electrophoresis (SDS-PAGE), Western blotting and radioimmunoprecipitation.
Materials and Methods Strains. The strains used were virulent and avirulent clones of L. interrogans serovar copenhageni, strain Shibaura (13, 17). The virulent clone was passaged in guinea pigs every 2 months and shown to be lethal for guinea pigs. The avirulent clone, derived from the virulent clone, was passaged in media for more than 15 years and determined to be avirulent. These strains were grown at 30°C in 0.2 % tryptose phosphate broth (Difco, Detroit, MI) containing 10% normal rabbit serum. Immune sera. Hyperimmune rabbit serum against the virulent and avirulent clones was prepared as previously described (16). The microscopic agglutination titres of the antivirulent clone antiserum and anti-avirulent clone antiserum were 1:102400 and 1:204800, respectively. Leptospiral extract for SDS-polyacrylamide gel electrophoresis (SDS-PA GE) and Western blotting. Leptospiral extracts were prepared as described by Yanagawa et al. (16) with a slight modification. Briefly, 4 X 1011 organisms were extracted with 1 ml of 0.5% SDS at 45°C for 4 h under shaking. Then, the extract was centrifuged at 15000 rpm on a Microfuge (Tomy Seiko, Tokyo) for 4 h, and the supernatant was used as the leptospiral extract sample. Absorption of anti-virulent clone antiserum with the avirulent clone. The anti-virulent clone antiserum was absorbed with the avirulent clone leptospiras as follows. 1 ml of avirulent clone leptospiras (1.0 X 1010 organisms/ml) suspended in phosphate-buffered
Leptospiral Antigens
455
saline (PBS, pH 7.4) was added to 25 Itl of anti-virulent clone antiserum, whose microscopic agglutination titre was 1:102400 in relation to the virulent clone. The mixture was incufor 1 h with occasional shaking and thereafter centrifuged. Further absorpbated at tion of the supernatant was carried out with 0.5 ml of the antigen. The absorption procedure was repeated until the supernatant did not agglutinate the avirulent clone at the dilution of 1:4, because of the high sensitivity of Western blotting. Ten ml of the leptospiral suspension was needed to absorb the anti-virulent clone antiserum to this level. Therefore, if we considered the dilution factor (25 1t11l0 ml = 400) alone, the titre of the original antivirulent clone antiserum to the virulent clone should be theoretically calculated as 1:256 (102400/400). Actually, after the absorption procedure, the titre of the antiserum to the virulent clone was 1:32. The reduction of the titre of the absorbed antiserum may have been due to factors of absorption. The same absorbed antiserum did not agglutinate the avirulent clone at a dilution of 1:4. The anti-avirulent clone antiserum which was absorbed with the virulent clone leptospiras showed no difference in the microscopic agglutination titres between the virulent clone and the avirulent clone. SDS-polyacrylamide gel electrophoresis (SDS-PAGE). Leptospiral extracts were separated on 12% polyacrylamide slab gels in the discontinuous Tris-glycine system as described by Laemmli (6). The samples were not boiled before electrophoresis. Gels were fixed and stained with Coomassie brilliant blue. The molecular weights were estimated by the method of Weber and Osborn (15). Western blotting. Western blotting was performed by the technique of Towbin et al. (12) with a Trans Blot Cell (Bio-Rad, Richmond, Ca.). Proteins were transferred to nitrocellulose membrane (Bio-Rad, Richmond, Ca.) at 15 rnA for 1 h following 30 rnA for 12 h with cooling in 25 mM Tris-192 mM glycine-12% (v/v) methanol buffer. The sheets were incubated with PBS (pH 7.4) containing 3% of bovine serum albumin (BSA, Boehringer Mannheim, Mannheim) for 1 h at room temperature to reduce nonspecific reaction. The sheets were incubated with anti-virulent clone or anti-avirulent clone antiserum at a dilution of 1/ 100 (or anti-virulent clone antiserum absorbed with avirulent clone at a dilution of 1/10) in 0.5% BSA-PBS for 2 h with constant rocking. Then, the sheets were rinsed with 0.05% Tween 20-PBS (PBST) and incubated with horseradish-peroxidase-conjugated anti-rabbit IgG (heavy and light chains, Bio-Rad, Richmond, Ca.) in 0.5% BSA-PBS at a dilution of 1/ 1000 for 2 h with constant rocking. The sheets were rinsed with PBSTand incubated with 4chloro-l-naphtol in 10 mM Tris-HCI buffer containing 100 mM NaCI (pH 8.0) for 10 to 20 min. Radiolabelling of leptospiras. Lactoperoxidase-catalyzed iodination of intact leptospiras was performed as described (1). Briefly, 0.5 mCi of Na 1251 (Amersham, Backinghamshire) and 100 Itl of lactoperoxidase (Sigma, St. Louis, MO.) were gently mixed with 0.5 ml of PBS containing 5.0 X 10 10 organisms. Then, 50 ul of H 202 solution (5 [!l of 30% stock H 202 freshly prepared in 125 ml of PBS) were induced, and the mixture was incubated for 15 min. Another 50 ul of H 2 0 2 solution were added and the reaction was allowed to stand for 15 min. The iodinated organisms were then washed and extracted with 0.5 ml of 0.5% SDS at 45 DC for 1 h. The extract was centrifuged at 15000 rpm on a Microfuge for 30 min, and the supernatant was used as radioiodinated leptospiral extract sample. Immunoprecipitation of radio iodinated leptospiral proteins. Immunoprecipitation was carried out by a modification of the procedure of Nunes-Edwards et al. (10). Biefly, 50 ul of the radioiodinated leptospiral extract were mixed with 450 [!lof Triton buffer (11) and then 30 ul of normal or test antisera were added. When the absorbed serum was used as the test serum, 90 Itl were added because the absorbed serum was rather low in its agglutination titre. After 1 h incubation at room temperature, 75 Itl of a slurry of protein A-Sepharose CL4B (Pharmacia, Uppsala) were added. The mixture was incubated at room temperature for 1 h with gentle shaking, pelleted and washed sufficiently with Triton buffer and then with 10 mM Tris-HCI buffer (pH 8.0). The immunoprecipitate was extracted from the pellet by adding 50 ul of SDS sample buffer (6) without 2-mercaptoethanol, heated in boiling water
3rc
456
M. Niikura, E. Ono, and R. Yanagawa
for 3 min and centrifuged in a Microfuge. A supernatant containing the precipitate was collected and analyzed by SDS-PAGE as previously described.
Results Comparison of protein profiles of virulent and avirulent clones by SDS-PAGE Protein profiles of virulent and avirulent clones of 1. interrogans serovar copenhageni, strain Shibaura were compared by SDS-PAGE (Fig. 1). More than 30 protein bands were detectable on either virulent or avirulent clones. The protein profiles of the virulent and avirulent clones were identical. Comparison of antigen profiles of virulent and avirulent clones by Western blotting 1) Comparison with anti-virulent clone antiserum. Anti-virulent clone antiserum was used first to identify the antigens of the virulent and avirulent clones (Fig. 2). More than 30 bands were visualized on the extracts from the virulent and avirulent clones,
MW _ 9 2 .5 K
-66 .2K
_45K
_31K
a
b
Fig. 1. Comparison of the protein profiles of the virulent (lane a) and avirulent (lane b) clones of 1. interrogans serovar copenhageni strain Shibaura. Samples were extracted with 0.5% SDS at 45°C for 4 hours and separated on 12% SDS-polyacrylamide gel. The gel was stained with Coomassie brilliant blue. No obvious differences in the position of bands are evident. The molecular weight markers appearing on the right represent the following protein standards: phosphorylase B, 92500; bovine serum albumin, 66200; ovalbumin, 45000; and carbonic anhydrase, 31000.
Leptospiral Antigens
457
MW
_92 .5K
_66.2K
a
b
c
d
Fig. 2. Comparison of antigen profiles of the virulent and avirulent clones by Western blotting using anti-virulent clone antiserum. Samples were extracted with 0.5% SDS at 45°C for 4 hours and separated on 12% SDS-polyacrylamide gel. Western blotting was performed as described in the text. Lanes: a. Virulent clone of L. interrogans serovar copenbageni strain Shibaura; b. Avirulent clone of L. interrogans serovar copenbageni strain Shibaura; c. Virulent clone of L. interrogans serovar copenhageni strain Shibaura heated at 100°C for 10 minutes before electrophoresis; d. Avirulent clone of L. interrogans serovar copenhageni strain Shibaura heated at 100°C for 10 minutes before electrophoresis. The bands which were different for the virulent and avirulent clones are indicated on the left. Molecular weight markers are given in the legend to Fig. 1.
respectively. Most of the antigen bands were identical, but there were two clear differences: 1) An antigen of about 32000 dalton was detected on the extract of the virulent clone but not on the extract of the avirulent clone; and 2) A 43000 dalton antigen was strongly stained only on the extract of the virulent clone, while the 41000 dalton antigen was strongly stained only on the avirulent clone. The 32000, 41000 and 43000 dalton antigens were not detected on the protein profiles on SDS-PAGE. 2) Comparison with anti-avirulent clone antiserum. Anti-avirulent clone antiserum was used to identify the antigens of the virulent and avirulent clones (Fig. 3). More than 30 bands were visualized on the extracts from the virulent and avirulent clones, respectively, by anti-avirulent clone antiserum as well as by anti-virulent clone antiserum. The 43000 dalton antigen on the virulent clone extract and the 41000 dalton
458
M. Niikura, E.Ono, and R. Yanagawa MW
_66.2K
--
_45K
_31K
a
b
c
d
Fig. 3 Comparison of antigen profiles of the virulent and avirulent clones by Western blotting using anti-avirulent clone antiserum. Samples was extracted with 0.5% SDS at 45°C for 4 hours and separated on 12% SDS-polyacrylamide gel. Western blotting was performed as described in the text. Lanes: a. Virulent clone of L. interrogans serovar copenhageni strain Shibaura; b. Avirulent clone of L. interrogans servar copenhageni strain Shibaura; c. Virulent clone of L. interrogans serovar copenhageni strain Shibaura heated at 100°C for 10 minutes before electrophoresis; d. Avirulent clone of L. interrogans serovar copenhageni strain Shibaura heated at 100°C for 10 minutes before electrophoresis. The bands which were different for the virulent and avirulent clones are indicated on the left. Molecular weight markers are given in the legend to Fig. 1. antigen on the avirulent clone extract were also detected with anti-avirulent clone antiserum. The 32000 dalton antigen found in the virulent clone extract by utilizing anti-virulent clone antiserum was not detected by utilizing the anti-avirulent clone antiserum. Heating the extract at 100°C for 10 min did not affect the antigen profiles of the extracts identified by both anti-virulent and avirulent clones antisera (Figs. 2 and 3, lanes c and d). Only a slight band of 45000 dalton was detected in the extracts of both clones with anti-virulent clone antiserum absorbed with the avirulent clone. The 32000 dalton antigen was not detected. No antigens were detected in the Western blotting experiments performed with normal rabbit serum as control. All the experiments using the Western blotting method were repeated more than three times and the same results were obtained.
Leptospiral Antigens
459
Comparison of surface-exposed antigenic protein by radioimmunoprecipitation using 125/ labelled leptospira 125
1 labelled organisms were examined by radioimmunoprecipitation to determine whether or not the 32000, 41000 and 43000 dalton antigens were surface-exposed proteins. As shown in Fig. 4, 16 proteins on the virulent clones extracts were radioiodinated (Fig. 4, lanes a and b). Both antisera precipitated 13 proteins of 93000, 76000, 66000, 62000, 58000, 46000, 40500, 36000, 33000, 29000, 23000, 18000 and MW
_
92 .5 K
_
66. 2 K
_45K
a
bed
e
f
9
h
Fig. 4. Comparison of surface-exposed antigenic proteins of the virulent and avirulent clones by radioimmunoprecipitation. Samples were SDS-soubilized 1251 labelled virulent (lanes a, c, e, g and i) and avirulent (b, d, f, hand j) clones of L. interrogans serovar copenhageni strain Shibaura. Lanes: a. 125I-labelled protein profile of the virulent clone; b. 125I-labelled protein profile of the avirulent clone; c. 12sI-labelled antigenic protein profile of the virulent clone precipitated with anti-virulent clone antiserum; d. 125I-labelled antigenic r:rotein profile of the avirulent clone precipitated with the anti-virulent clone antiserum; e. 2sI-labelled antigenic protein profile of the virulent clone precipitated with the anti-avirulent clone antiserum; f. 12sI-labelled antigenic protein profile of the avirulent clone precipitated with the anti-avirulent clone antiserum; g. 125I-labelled protein profile of the virulent clone precipitated with the anti-virulent clone antiserum absorbed with the avirulent clone; h. 12sl-labelled protein profile of the avirulent clone precipitated with anti-virulent clone antiserum absorbed with the avirulent clone; i. WI-labelled protein profile of the virulent clone precipitated with normal rabbit serum; j. WI-labelled protein profile of the avirulent clone precipitated with normal rabbit serum. No obvious differences are evident among lanes a, b, c and d. Molecular weight markers are given in the legend to Fig. 1.
460
M. Niikura, E. Ono, and R. Yanagawa
16000 dalton on the extracts of both virulent and avirulent clones (Fig. 4, lanes c and d; e and f). Of these proteins, however, 2 proteins of 93000 and 16000 dalton were precipitated with normal rabbit serum (Fig. 4, lanes i and i), indicating that these 2 proteins were precipitated non-specifically. The 32000, 41000 and 43000 dalton antigens were not precipitated with the anti-virulent and anti-avirulent clones antisera, suggesting that these antigens were not surface-exposed proteins. Anti-virulent clone antiserum absorbed with the avirulent clone precipitated 2 proteins, which were identical to those precipitated with normal rabbit serum, showing that the 2 proteins were non-specifically precipitated (Fig. 4, lanes g and h). This fact may indicate that the so called "Vi antigen", which has been thought to be present on leptospiral cells of the virulent clone and was detectable by the microscopic agglutination test, was not demonstrated by radioimmunoprecipitation.
Discussion The antigen specific to the virulent clone of L. interrogans serovar copenhageni, strain Shibaura was detected in the present study by Western blotting. The 32000 dalton antigen specific to the virulent clone was thought to be of non-protein nature because no corresponding band was detected on the protein profiles on SDS-PAGE stained with Coomassie brilliant blue. This antigen was considered not to be a surface antigen protein because no corresponding band was detected by radioimmunoprecipitation, A lipopolysaccharide in Haemophilus influenzae is a precedential example of a non-protein antigen, which was detected by Western blotting (7). The chemical nature of the 32000 dalton antigen will be a subject of future study. The 32000 dalton antigen was not detected, contrary to expectation, by Western blotting with the absorbed serum. One explanation may be as follows. The 32000 dalton antigen found only in the virulent clone (but not large in amount) may be present in the avirulent clone in such a small amount as to be undetectable by Western blotting and the antibodies to the antigen may have been absorbed during the absorption procedure. The 43000 dalton antigen molecule was detected only in the virulent clone, while the 41000 dalton antigen molecule was detected only in the avirulent clone. These antigenic molecules were, however, detected with both anti-virulent and anti-avirulent clone antisera. These antigenic molecules of the virulent and avirulent clones may share at least one common antigenic determinant, but they migrated to different positions on SDS-PAGE. Antigens which have a several common antigenic determinant(s) and different molecular weight were reported to exist in strains of Haemophilus influenzae (4). The 43000 and 41000 dalton antigen molecules detected by Western blotting, however, had no corresponding bands on either SDS-PAGE stained with Coomassie brilliant blue or radioimmunoprecipitation using 1251 labelling. These antigen molecules were therefore considered to be non-protein antigens. The 32000 dalton antigen specific to the virulent clone was found to be thermostable. The 32000 dalton antigen may, therefore, be different from the "Vi antigen". Ueno et al. (14) reported that the thermolabile "Vi antigen" was found in the virulent clone of L. tnterrogans serovar copenhageni, strain Shibaura by the microscopic agglutination test with the anti-virulent clone antiserum absorbed with the avirulent clone organisms. The agglutination titre of "Vi antigen" detected by the absorbed serum was rather low. In the present study, "Vi antigen" was not detected by Western
Leptospiral Antigens
461
blotting, but detected in the microscopic agglutination test. The fact that the thermolabile antigen was detected by the microscopic agglutination test agreed with a number of reports showing that leptospiras have thermolabile antigens (2, 8, 9). The reasons why "Vi antigen" was not detected by Western blotting is not clear. However, some possible reasons why "Vi antigen" was not detected by Western blotting could be: 1) Thermolabile and thermostable antigenic determinants may coexist on an antigen molecule; 2) Thermolabile and thermostable antigens may have comigrated to the same position in SDS-PAGE; and 3) Several antigenic molecules including thermolabile antigen may not have been transferred to the nitrocellulose sheet. Differences in the colony type, susceptibility to leptospiricidal activity mediated by antibody plus complement (17) and ability to attach to cultured cells (13) were reported to exist between the virulent and avirulent clones. How the 32000, 41000 and 43000 dalton antigens, which were different in the virulent and avirulent clones, correlate to such properties and to virulence is another subject for future study. Acknowledgments. This work was supported by Grants-in-Aid for Scientific Research Nos. 60440020 and 60760221 from the Ministry of Eduation, Science and Culture of Japan. All experiments utilizing radioisotopes could be carried out at the Central Institute of Radioisotope Science, Hokkaido University, through the cooperation kindly provided by Prof. S. Nishi and Dr. T. Ohnishi.
References
1. Alderete, ]. F. and]. B. Baseman: Surface characterization of virulent Treponema pallidum. Infect. Immun. 30 (1980) 814-823 2. Borg-Petersen, c.: A thermo-labile antigen in the leptospira strain Ictero No. 1. Trop. geogr. Med. 23 (1971) 282-285 3. Paine, S.: Leptospirosis. 37-98. In: Guidelines for the control of leptospirosis. Ed. S. Faine World Health Organization, Geneva (1982) 4. Gulig, P. A., G. H. McCracken jr., c. F. Frisch, K. H. Johnson, and E. ]. Hansen: Antibody response of infants to cell surface exposed outer membrane protein of Haemophilus influenzae Type B after systemic Haemophilus disease. Infect. Immun. 37 (1982) 82-88 5. Kmety, E.: Ober ein Vi-Antigen bei Leptospiren. Zbl. Bakt. Hyg., I. Abt, Orig. A 211 (1972) 343-351 6. Laemmli, U. K.: Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227 (1970) 680-685 7. Loeb, M. R.: Immunoblot method for identifying components, determinating their cross-reactivity, and investigating cell topology: Results with Haemophilus influenzae Type B. Analyt, Biochem. 143 (1984) 196-204 8. Manev, c.: Serological characterization of the leptospira serogroup pomona. I. Factor analysis of the reference strains. Zbl. Bakt. Hyg., I. Abt. Orig. A 236 (1976) 316-322 9. Nicolescu, M. and G H. Moldoveanu: The thermolabile antigen of some reference leptospira strains from the pomona serogroup. Arch. Roum. Path. Exp. Microbiol. 36 (1977) 151-155 10. Nunes-Edwards, P. L., A. B. Thiermann, P.]. Bassford [r., and L. V. Stamm: Identification and characterization of the protein antigen of Leptospira interrogans serovar hard;0. Infect. 1mmun. 48 (1985) 482-497
462
M. Niikura, E.Ono, and R. Yanagawa
11. Stamm, L. V., T. C. Kerner jr., V. A. Bankaitis, and P. J. Bassford jr.: Identification and preliminary characterization of Treponema pallidum protein antigens expressed in Escherichia coli. Infect. Immun. 41 (1983) 709-721 12. Toiobin, H., T. Staehelin, and J. Gordon: Electrophoretic transfer of protein from polyacrylamide gel to nitrocellulose sheets: Procedure and some applications. Proc. Nat. Acad. Sci. USA 76 (1979) 4350-4354 13. Tsuchimoto, M., M. Niikura, E. Ono, and R. Yanagawa: Leptospiral attachment to cultured cells. Zbl. Bakt. Hyg., I. Abt. Orig. A 258 (1984) 268-274 14. Ueno, K., R. Yanagaioa, and H. Kida: Presence of Vi antigen in a virulent strain of Leptospira interrogans serovar pomona and relation of Vi antigens of leptospira to resistance to leptospiricidal activity mediated by antiserum plus complement. Zbl. Bakt. Hyg., I. Abt, Orig. A 252 (1982) 557-565 15. Weber, K. and M. Osborn: The reliability of molecular weight determinations by dodecyl sulfate-polyacrylamide gel electrophoresis. J. BioI. Chern. 244 (1964) 4406-4412 16. Yanagauia, R., M. Shinagawa, and I. Takashima: Serological studies of leptospires by immunodiffusion. Zbl. Bakt. Hyg., I. Abt. Orig. A 228 (1974) 369-377 17. Yepez-Plascencia, W., R. Yanagauia, and K. Ueno: Correlation of virulence, susceptibility to leptospiricidal activity test mediated by antiserum plus complement and colonial morphology of five lines of a strain of Leptospira interrogans serovar copenhageni. Zbl. Bakt. Hyg., I. Abt, Orig. A 251 (1981) 230-256 Professor Dr. Ryo Yanagatoa, Hokkaido University, Faculty of Veterinary Medicine, Dept. of Hygiene and Microbiology, Sapporo 060, Japan