- Email: [email protected]
Species differentiation of Leptospira interrogans serovar hardjo strain Hardjobovis from strain Hardjoprajitno by dna slot blot hybridisation
Research in Veterinary Science 1990, 49, 194-197
Species differentiation of Leptospira interrogans serovar hardjo strain Hardjobovis from strain Hardjoprajitno by DNA slot blot hybridisation P. RAMADASS, R. B. MARSHALL, Department of Veterinary Pathology and Public Health, Massey University, Palmerston North, New Zealand, B. D. W. JARVIS, Department ofMicrobiology and Genetics, Massey University, Palmerston North, New Zealand
have described five new species among the parasitic Leptospira based on DNA homology groups. The present study was undertaken to examine differences between strains Hardjobovis and Hardjoprajitno by DNA-DNA hybridisation to determine the DNA homology groups to which they belong.
Slot blot hybridisation studies with total genomic DNA probes were used to compare Leptospira tnterrogans serovar hardjo strain Hardjoprajitno, strain Hardjobovis and a number of other Leptospira interrogans serovars, Strains Hardjoprajitno and Hardjobovis were found to have little genetic relationship with each other when compared to some the other serovars tested. Hardjoprajitno is closely related to serovar icterohaemorrhagiae and not to Hardjobovis whereas Hardjobovis is closely related to serovars vietnam, balcanica andjavanica but not to serovar icterohaemorrhagiae; this places strain Hardjoprajitno in the species L interrogans and strain Hardjobovis in the species L borgpetersoni, Because of this lack of genetic relatedness between strains Hardjoprajitno and Hardjobovis, it is proposed to remove the prefix Hardjo from the strain name Hardjobovis and call it L borgpetersoni serovar hardjo strain Bovis.
of
Materials and methods Leptospira serovars and culture conditions
BRITISH, North American, Australian and New Zealand cattle are commonly infected with Leptospira interrogans serovar hardjo. For many years it was thought that a single strain was involved and that this was similar, if not identical, to the reference strain Hardjoprajitno. The first real indication that this was not the case came from New Zealand studies (Robinson et al 1982), using DNA restriction endonuclease analysis. These studies compared cattle isolates with the reference strain and showed marked differences in the electrophoretic patterns of their genomic DNA. Similar observations were also made with isolates from North America (Thier mann et al 1986, Le Febvre and Thiermann 1986) and the United Kingdom (Marshall et al 1985). The genetically distinct subtype, that was first recognised by this technique, was called Hardjobovis (Marshall et al 1985, Zuerner and Bolin 1988). Both Hardjobovis and Hardjoprajitno strains have been identified among cattle isolates from the United Kingdom (Marshall et al 1985). Recently, Yasuda et al (1987)
The L interrogans serovar hardjo strain Hardjoprajitno used in this work was the reference strain maintained in our laboratory (Marshall et al 1981, Robinson et al 1982). Leptospira borgpetersoni serovar hardjo strain Hardjobovis was isolated from a field case of the infection in a cow. They were grown in Ellinghausen, McCullough, Johnson and Harris (EMIH) medium (Johnson and Harris 1967) (Difco Laboratories) for five to seven days at 30°C. Serovar balcanica (strain T78) was isolated from a possum (Trichosurus vulpecula) in New Zealand (Marshall et a11976) and has since been maintained in this laboratory. All other strains used were originally obtained from the WHO/FAO Collaborating Leptospirosis Laboratory, State Health Department, Queensland, Australia (Table 1).
Preparation
0/ DNA
samples
The leptospiral DNA was prepared from 1 litre cultures in EMJH medium by the method of Marshall et al (1981) using the phenolchloroform-isoamyl alcohol extraction procedure. The DNA was further purified by caesium chloride density gradient centrifugation (Maniatis et al 1982).
Restriction endonuclease analysis The leptospiral DNA was digested with enzyme EcoRl, separated by electrophoresis and photographed by the methods of Marshall et al (1981).
194
Differentiation of Leptospira interrogans serovars TABLE 1: List of serovars used Serogroup
Serovar
Strain
Sejroe Ballum Cynopteri Cellodoni Mini Sejroe Sejroe
balcanica bal/um cynopteri eel/odoni georgia hardjo hardjo
T 78 Mus 127 3522 C Cellodoni LT 117 Hardjoprajitno Hardjobovis (field strain)
ieterohaemorrhagiae javaniea kabura kisuba mini moldaviae* pomona vietnam *
RGA Veldrat Batavia 46 Kabura Kisuba Sari
or~agiae
Icterohaem Javanica Hebdomadis Tarassovi Mini Tarassovi Pomona Uncertain
114-2 Pomona Vietnam
* These two serovars are not included in the revised list of Leptospira serovars by E. Kmety and H. Dikken, University Press, Groningen, July 1988
Slot blot hybridisation Slot blot hybridisation was performed according to the method of Kafatos et al (1979) with some modifications. DNA samples (0· 5 Jlg) in 10 Jll of deionised water were denatured with 100 III of 0·4 M sodium hydroxide solution for 10 minutes at room temperature, neutralised with 100 ILl of 2 M ammonium acetate solution and applied to a nitrocelluluse membrane (BA 85 Schleicher and Schuell, Dassel) in a slot blot apparatus (Bio Rad) under vacuum. The nitrocellulose membrane was placed in a vacuum oven at 80°C for two hours before hybridisation. The filter membrane was then prehybridised with about 10 ml of hybridisation buffer (Jarvis et al 1983) in a sealed plastic bag at' 60°C in a water bath for two hours. Leptospiral DNA (3 p.g)was radiolabelled following sonication with 32P ..labelled deoxyadenosine triphosphate by the random primer method (Taylor et al 1976), heat denatured and added (1 to 1· 5 X 106 counts per minute) to the plastic bag containing the prehybridised filter membrane and hybridisation buffer. DNA was allowed to reanneal for 24 hours at 60°C in a waterbath. Then the filter membrane was removed, washed twice with 2 x sse (1 x sse is O· 15 M sodium chloride plus 0-015 M sodium citrate) for 15 minutes at room temperature, once in 0·1 x sse for 15 minutes at room temperature and 'once in 0·1 x sse for 15 minutes at 50°C in a waterbath. The filter membrane was dried at room temperature, wrapped in plastic cling film and exposed overnight to .X-ray film (Agfa-Gevaert, Curix RP 2) using an intensifying screen(Dupont Cronex Lighting Plus) at -70°C. The autoradiogram was developed in a Kodak automatic film processor. A quantitative estimate of DNA homology was obtained by cutting individual slots from the filter membrane and count-
195
ing the associated radioactivity in a liquid scintillation counter (Seldin and Dubnau 1985). The percentage of homology was calculated as follows Counts per minute with heterologous Percentage homology
DNA
=C t. t oun s per mmu e
X
100.
with homologous DNA
Results Slot blot hybridisation results are shown in Figs 1 and 2 as autoradiograms. In Fig I the filter was probed with DNA from strain Hardjobovis and in Fig 2 with DNA from strain Hardjoprajitno. When the Hardjobovis probe was used (Fig 1), serovars balcanica and vietnam showed 100 per cent homology; javanica 81 per cent and moldaviae 78 per cent indicating that they all belong to the same genetic homology group. Hardjoprajitno DNA showed 2-8 per cent homology (row 2, A and B). Serovars cynopteri and cel/edoni showed 3· 4 per cent and 23 per cent homology, respectively. In autoradiograms not shown, this probe also indicated a relationship with tarassovi but not with ballum, When Hardjoprajitno DNA was used as the probe (Fig 2), serovars icterohaemorrhagiae showed 100 per cent homology, indicating that it belongs to the same homology group. Hardjobovis DNA showed 8 per cent homology, georgia 3 per cent, kisuba 4 per cent, mini 3 per cent and kabura 37 per cent homology. The Escherichia coli DNA which was used as a negative control gave 0 per cent homology. In autoradiograms not shown the Hardjoprajitno probe indicated a relationship with pomona but not tarassovi or ballum.
t.~,:~i.~.~.73:
1 2 f: ~"''''h
ri"i 4
~~~
.
D
ABC ,J"' "
~nt'•• ·:",',:\"·~",• ..,.::
.-..-".
"""
..
,~ "~' .,~
,"::
~~:~, :C~"
FIG 1: Autoradiogram of a slot blot hybridisation between DNA from various Leptospira serovars and 32P-labelled total genomic DNA from Leptospira bor9petersoni serovar hardjo strain Hardjobovis. Unlabelled DNA samples are spotted in duplicates. 1, A and B strain Hardjoprajitno; 1, C and 0 serovar balcanica; 2, A and B serovar cynopteri; 2, C and 0 serovar cel/edoni; 3, A and B serovar javaniea; 3, C and 0 serovar mokieviee: 4, A and B serovar vietnam; 4, C and 0 strain Hardjobovis
P. Ramadass, R. B. Marshall, B. D. W. Jarvis
196
B
4
5
7.";;!,.: 8
•.
.~
,.
FIG 2: Autoradiogram of a slot blot hybridisation between DNA from various Leptospira serovars and 32pMlabelled total genomic DNA from Leptospira interrogans serovar hardjo strain Hardjoprajitno. A 1 and 2 strain Hardjoprajitno; A 3 and 4 serovar georgia; A 5 and 6 E coli; A 7 and 8 serovar icterohaemorrhagiae; Bland 2 serova r kisuba; B 3 and 4 serov ar mini; B 5 and 6 serovar kabura and B 7 and 8 strain Hardjobovis
Discussion In recent years, two distinct genotypes of serovar hardjo have been reported from various geographical regions of the world (Marshall et al 1985, Ellis et al 1986, Le Febvre and Thiermann 1986, Zuerner and Bolin 1988)and a difference in their pathogenicity has been suggested (Marshall et al 1985, Ellis et al 1988, Le Febvre et al 1987). Therefore, the detection of the genetic type of serovar hardjo which occurs in a particular locality, may be important with respect to epidemiological studies and the production of vaccines. Restriction endonuclease analysis of total genomic DNA will clearly differentiate strains Hardjoprajitno and Hardjobovis from one another (Robin-
son et al 1982, Ellis et al 1988). This difference in restriction pattern does not, however, indicate the extent to which these two strains differ genetically. The G+ C content of the DNA from strain Hardjoprajitno is 34 to 35 per cent compared with Hardjobovis which is 38 to 40 per cent (Le Febvre et aI1987). Southern blots of total genomic DNA demonstrate that Hardjoprajitno and Hardjobovis have little genetic homology (Le Febvre and Thiermann 1986). More recently, DNA clones were prepared from strain Hardjobovis and used as probes in hybridisation experiments to differentiate strain Hardjobovis from strain Hardjoprajitno (Van Eys et al 1988, Zuerner and Bolin 1988). These workers found that total DNA probes were not specific enough to differentiate between different strains. This, however, was not the present authors' experience and in this study the total genomic probe is easily able to differentiate these two strains (Figs 1 and 2). There is a general move in bacterial taxonomy towards classification based on genetic relatedness. The work of Yasuda et al (1987) is a major contribution preparing for such a move within the genus Leptospira. The close relatedness of strain Hardjoprajitno to serovar icterohaemorrhagiae (Fig 1) and pomona indicates that this strain belongs to the species which Yasuda et al (1987) have named L interrogans. On the other hand, strain Hardjobovis shows little relationship with serovar icterohaemorrhagiae but shows a high degree of relatedness to serovars balcanica and [avanica. This places strain Hardjobovis in the new species, L borgpetersoni, proposed by Yasuda et al (1987). It is perhaps surprising to find that two strains which belong to the same serovar (hardjo) are so genetically distinct as to belong in separate species. This finding indicates that these strains represent genetically distinct populations' which have evolved antigenic determinants indistinguishable by accepted methods for serovar determination. It may not be a coincidence that these two antigenically similar strains are also both parasites well adapted to cattle. The genetic similarities between serovar balcanica and strain Hardjobovis as judged by G + C content (Le Febvre et al 1987) and DNA hybridisation (Fig 1) indicate that they have a much closer evolutionary relationship to one another than do strains Hardjobovis and Hardjoprajitno. In the future, serovar determination studies which have formed the basis of the present classification scheme will probably be more appropriately used for subspecies differentiation. With the advent of a Leptospira classification scheme based primarily upon genetic relatedness, the two strains under study should be placed in different species. It is therefore proposed in this publication to drop the prefix Hardjo from the strain name Hardjo..
Differentiation of Leptospira interrogans serovars bovis and call the organism L borgpetersoni serovar Hardjo strain Bovis. Acknowledgements This work was financed by a grant from Veterinary Public Health, Division of Communicable Diseases, WHO (Geneva) and Massey University Research Fund.
"
References ELLIS, W. A., SONGER, J. G., MONTGOMERY, J. s CASSELLS, J. AS. (1986) Veterinary Record 118, 11-13 ELLIS, W. A., THIERMANN, A. B., MONTGOMERY, J., HANDSAKER, A., WINTER, P. J. & MARSHALL, R. B. (1988) Research in Veterinary Science 44, 375-379 JARVIS, B. D. W., SCOTT, K. F., HUGHES, J. E., DJORD~ JEVIe, N., ROLFE, B: G. & SHINE, J. (1983) Canadian Journal of Microbiology 29,200-209 JOHNSON, R. D. & HARRIS, V. G. (1967) Journal ofBacteriology 94,27-31 KAFATOS, F. C., JONES, C. W. & EFSTRADIADIS, A. (1979) Nucleic Acid Research 7, 1541-1552 LE FEBVRE, R. B. & THIERMANN, A. B. (1986) American Journal of Veterinary Research 47, 959-963 LE FEBVRE, R. B., THIERMANN, A. B. & FOLEY, J. (1987) Journal of Clinical Microbiology 25, 2094-2097
.
197
MANIATIS, T., FRITSCH, E. F. & SA~IBROOK, J. (1982) Molecular Cloning: a Laboratory Manual. New York, Cold Spring Harbor Laboratory MARSHALL, R. B., MANKTELOW, B. W., RYAN, T. J. & HATHAWAY, S. C. (1976) New Zealand Medical Journal 84, 74-75 MARSHALL, R. B., WILTON, B. E. & ROBINSON, A. J. (1981) Journal of Medical Microbiology 14, 163-166 MARSHALL, R. B., WINTER, P. J., THIERMANN, A. B. & ELLIS, W. A. (1985) Veterinary Record 117,669-670 ROBINSON, A. J., RAMADASS, P., LEE, A. & MARSHALL, R. B. (1982) Journal of Medical Microbiology 15,331-338 SELDIN, L. & DUBNAU, D. (1985) International Journal of Systematic Bacteriology 35, 151-154 TAYLOR, J. M., ILLMENSEE, R. & SUMMERS, J. (1976) Biochimica et Biophysica Acta 442,324-330 THIERMANN, A. B., HANDSAKER, A. L., FOLEY, J. W., WHITE, F. H. & KINGSCOTE, B. F. (1986) American Journal of Veterinary Research 47, 61-66 VAN EYS, G. J. J. M., ZAAL, J., SCHOONE, G. J. & TERPSTRA, W. J. (1988) Journal of General Microbiology 134, 567-574 YASUDA, P. H., STEIGERWALT, A. G., SULZER, K. R., KAUFMANN, A. F., ROGERS, F. & BRENNER, D. J. (1987) International Journal of Systematic Bacteriology 37,407-415 ZUERNER, R. & BOLIN, C. (1988) American Society for Microbiology D66, 861
Received Apri/2/, /989 Accepted February /9, /990
Species differentiation of Leptospira interrogans serovar hardjo strain Hardjobovis from strain Hardjoprajitno by DNA slot blot hybridisation P. RAMADASS, R. B. MARSHALL, Department of Veterinary Pathology and Public Health, Massey University, Palmerston North, New Zealand, B. D. W. JARVIS, Department ofMicrobiology and Genetics, Massey University, Palmerston North, New Zealand
have described five new species among the parasitic Leptospira based on DNA homology groups. The present study was undertaken to examine differences between strains Hardjobovis and Hardjoprajitno by DNA-DNA hybridisation to determine the DNA homology groups to which they belong.
Slot blot hybridisation studies with total genomic DNA probes were used to compare Leptospira tnterrogans serovar hardjo strain Hardjoprajitno, strain Hardjobovis and a number of other Leptospira interrogans serovars, Strains Hardjoprajitno and Hardjobovis were found to have little genetic relationship with each other when compared to some the other serovars tested. Hardjoprajitno is closely related to serovar icterohaemorrhagiae and not to Hardjobovis whereas Hardjobovis is closely related to serovars vietnam, balcanica andjavanica but not to serovar icterohaemorrhagiae; this places strain Hardjoprajitno in the species L interrogans and strain Hardjobovis in the species L borgpetersoni, Because of this lack of genetic relatedness between strains Hardjoprajitno and Hardjobovis, it is proposed to remove the prefix Hardjo from the strain name Hardjobovis and call it L borgpetersoni serovar hardjo strain Bovis.
of
Materials and methods Leptospira serovars and culture conditions
BRITISH, North American, Australian and New Zealand cattle are commonly infected with Leptospira interrogans serovar hardjo. For many years it was thought that a single strain was involved and that this was similar, if not identical, to the reference strain Hardjoprajitno. The first real indication that this was not the case came from New Zealand studies (Robinson et al 1982), using DNA restriction endonuclease analysis. These studies compared cattle isolates with the reference strain and showed marked differences in the electrophoretic patterns of their genomic DNA. Similar observations were also made with isolates from North America (Thier mann et al 1986, Le Febvre and Thiermann 1986) and the United Kingdom (Marshall et al 1985). The genetically distinct subtype, that was first recognised by this technique, was called Hardjobovis (Marshall et al 1985, Zuerner and Bolin 1988). Both Hardjobovis and Hardjoprajitno strains have been identified among cattle isolates from the United Kingdom (Marshall et al 1985). Recently, Yasuda et al (1987)
The L interrogans serovar hardjo strain Hardjoprajitno used in this work was the reference strain maintained in our laboratory (Marshall et al 1981, Robinson et al 1982). Leptospira borgpetersoni serovar hardjo strain Hardjobovis was isolated from a field case of the infection in a cow. They were grown in Ellinghausen, McCullough, Johnson and Harris (EMIH) medium (Johnson and Harris 1967) (Difco Laboratories) for five to seven days at 30°C. Serovar balcanica (strain T78) was isolated from a possum (Trichosurus vulpecula) in New Zealand (Marshall et a11976) and has since been maintained in this laboratory. All other strains used were originally obtained from the WHO/FAO Collaborating Leptospirosis Laboratory, State Health Department, Queensland, Australia (Table 1).
Preparation
0/ DNA
samples
The leptospiral DNA was prepared from 1 litre cultures in EMJH medium by the method of Marshall et al (1981) using the phenolchloroform-isoamyl alcohol extraction procedure. The DNA was further purified by caesium chloride density gradient centrifugation (Maniatis et al 1982).
Restriction endonuclease analysis The leptospiral DNA was digested with enzyme EcoRl, separated by electrophoresis and photographed by the methods of Marshall et al (1981).
194
Differentiation of Leptospira interrogans serovars TABLE 1: List of serovars used Serogroup
Serovar
Strain
Sejroe Ballum Cynopteri Cellodoni Mini Sejroe Sejroe
balcanica bal/um cynopteri eel/odoni georgia hardjo hardjo
T 78 Mus 127 3522 C Cellodoni LT 117 Hardjoprajitno Hardjobovis (field strain)
ieterohaemorrhagiae javaniea kabura kisuba mini moldaviae* pomona vietnam *
RGA Veldrat Batavia 46 Kabura Kisuba Sari
or~agiae
Icterohaem Javanica Hebdomadis Tarassovi Mini Tarassovi Pomona Uncertain
114-2 Pomona Vietnam
* These two serovars are not included in the revised list of Leptospira serovars by E. Kmety and H. Dikken, University Press, Groningen, July 1988
Slot blot hybridisation Slot blot hybridisation was performed according to the method of Kafatos et al (1979) with some modifications. DNA samples (0· 5 Jlg) in 10 Jll of deionised water were denatured with 100 III of 0·4 M sodium hydroxide solution for 10 minutes at room temperature, neutralised with 100 ILl of 2 M ammonium acetate solution and applied to a nitrocelluluse membrane (BA 85 Schleicher and Schuell, Dassel) in a slot blot apparatus (Bio Rad) under vacuum. The nitrocellulose membrane was placed in a vacuum oven at 80°C for two hours before hybridisation. The filter membrane was then prehybridised with about 10 ml of hybridisation buffer (Jarvis et al 1983) in a sealed plastic bag at' 60°C in a water bath for two hours. Leptospiral DNA (3 p.g)was radiolabelled following sonication with 32P ..labelled deoxyadenosine triphosphate by the random primer method (Taylor et al 1976), heat denatured and added (1 to 1· 5 X 106 counts per minute) to the plastic bag containing the prehybridised filter membrane and hybridisation buffer. DNA was allowed to reanneal for 24 hours at 60°C in a waterbath. Then the filter membrane was removed, washed twice with 2 x sse (1 x sse is O· 15 M sodium chloride plus 0-015 M sodium citrate) for 15 minutes at room temperature, once in 0·1 x sse for 15 minutes at room temperature and 'once in 0·1 x sse for 15 minutes at 50°C in a waterbath. The filter membrane was dried at room temperature, wrapped in plastic cling film and exposed overnight to .X-ray film (Agfa-Gevaert, Curix RP 2) using an intensifying screen(Dupont Cronex Lighting Plus) at -70°C. The autoradiogram was developed in a Kodak automatic film processor. A quantitative estimate of DNA homology was obtained by cutting individual slots from the filter membrane and count-
195
ing the associated radioactivity in a liquid scintillation counter (Seldin and Dubnau 1985). The percentage of homology was calculated as follows Counts per minute with heterologous Percentage homology
DNA
=C t. t oun s per mmu e
X
100.
with homologous DNA
Results Slot blot hybridisation results are shown in Figs 1 and 2 as autoradiograms. In Fig I the filter was probed with DNA from strain Hardjobovis and in Fig 2 with DNA from strain Hardjoprajitno. When the Hardjobovis probe was used (Fig 1), serovars balcanica and vietnam showed 100 per cent homology; javanica 81 per cent and moldaviae 78 per cent indicating that they all belong to the same genetic homology group. Hardjoprajitno DNA showed 2-8 per cent homology (row 2, A and B). Serovars cynopteri and cel/edoni showed 3· 4 per cent and 23 per cent homology, respectively. In autoradiograms not shown, this probe also indicated a relationship with tarassovi but not with ballum, When Hardjoprajitno DNA was used as the probe (Fig 2), serovars icterohaemorrhagiae showed 100 per cent homology, indicating that it belongs to the same homology group. Hardjobovis DNA showed 8 per cent homology, georgia 3 per cent, kisuba 4 per cent, mini 3 per cent and kabura 37 per cent homology. The Escherichia coli DNA which was used as a negative control gave 0 per cent homology. In autoradiograms not shown the Hardjoprajitno probe indicated a relationship with pomona but not tarassovi or ballum.
t.~,:~i.~.~.73:
1 2 f: ~"''''h
ri"i 4
~~~
.
D
ABC ,J"' "
~nt'•• ·:",',:\"·~",• ..,.::
.-..-".
"""
..
,~ "~' .,~
,"::
~~:~, :C~"
FIG 1: Autoradiogram of a slot blot hybridisation between DNA from various Leptospira serovars and 32P-labelled total genomic DNA from Leptospira bor9petersoni serovar hardjo strain Hardjobovis. Unlabelled DNA samples are spotted in duplicates. 1, A and B strain Hardjoprajitno; 1, C and 0 serovar balcanica; 2, A and B serovar cynopteri; 2, C and 0 serovar cel/edoni; 3, A and B serovar javaniea; 3, C and 0 serovar mokieviee: 4, A and B serovar vietnam; 4, C and 0 strain Hardjobovis
P. Ramadass, R. B. Marshall, B. D. W. Jarvis
196
B
4
5
7.";;!,.: 8
•.
.~
,.
FIG 2: Autoradiogram of a slot blot hybridisation between DNA from various Leptospira serovars and 32pMlabelled total genomic DNA from Leptospira interrogans serovar hardjo strain Hardjoprajitno. A 1 and 2 strain Hardjoprajitno; A 3 and 4 serovar georgia; A 5 and 6 E coli; A 7 and 8 serovar icterohaemorrhagiae; Bland 2 serova r kisuba; B 3 and 4 serov ar mini; B 5 and 6 serovar kabura and B 7 and 8 strain Hardjobovis
Discussion In recent years, two distinct genotypes of serovar hardjo have been reported from various geographical regions of the world (Marshall et al 1985, Ellis et al 1986, Le Febvre and Thiermann 1986, Zuerner and Bolin 1988)and a difference in their pathogenicity has been suggested (Marshall et al 1985, Ellis et al 1988, Le Febvre et al 1987). Therefore, the detection of the genetic type of serovar hardjo which occurs in a particular locality, may be important with respect to epidemiological studies and the production of vaccines. Restriction endonuclease analysis of total genomic DNA will clearly differentiate strains Hardjoprajitno and Hardjobovis from one another (Robin-
son et al 1982, Ellis et al 1988). This difference in restriction pattern does not, however, indicate the extent to which these two strains differ genetically. The G+ C content of the DNA from strain Hardjoprajitno is 34 to 35 per cent compared with Hardjobovis which is 38 to 40 per cent (Le Febvre et aI1987). Southern blots of total genomic DNA demonstrate that Hardjoprajitno and Hardjobovis have little genetic homology (Le Febvre and Thiermann 1986). More recently, DNA clones were prepared from strain Hardjobovis and used as probes in hybridisation experiments to differentiate strain Hardjobovis from strain Hardjoprajitno (Van Eys et al 1988, Zuerner and Bolin 1988). These workers found that total DNA probes were not specific enough to differentiate between different strains. This, however, was not the present authors' experience and in this study the total genomic probe is easily able to differentiate these two strains (Figs 1 and 2). There is a general move in bacterial taxonomy towards classification based on genetic relatedness. The work of Yasuda et al (1987) is a major contribution preparing for such a move within the genus Leptospira. The close relatedness of strain Hardjoprajitno to serovar icterohaemorrhagiae (Fig 1) and pomona indicates that this strain belongs to the species which Yasuda et al (1987) have named L interrogans. On the other hand, strain Hardjobovis shows little relationship with serovar icterohaemorrhagiae but shows a high degree of relatedness to serovars balcanica and [avanica. This places strain Hardjobovis in the new species, L borgpetersoni, proposed by Yasuda et al (1987). It is perhaps surprising to find that two strains which belong to the same serovar (hardjo) are so genetically distinct as to belong in separate species. This finding indicates that these strains represent genetically distinct populations' which have evolved antigenic determinants indistinguishable by accepted methods for serovar determination. It may not be a coincidence that these two antigenically similar strains are also both parasites well adapted to cattle. The genetic similarities between serovar balcanica and strain Hardjobovis as judged by G + C content (Le Febvre et al 1987) and DNA hybridisation (Fig 1) indicate that they have a much closer evolutionary relationship to one another than do strains Hardjobovis and Hardjoprajitno. In the future, serovar determination studies which have formed the basis of the present classification scheme will probably be more appropriately used for subspecies differentiation. With the advent of a Leptospira classification scheme based primarily upon genetic relatedness, the two strains under study should be placed in different species. It is therefore proposed in this publication to drop the prefix Hardjo from the strain name Hardjo..
Differentiation of Leptospira interrogans serovars bovis and call the organism L borgpetersoni serovar Hardjo strain Bovis. Acknowledgements This work was financed by a grant from Veterinary Public Health, Division of Communicable Diseases, WHO (Geneva) and Massey University Research Fund.
"
References ELLIS, W. A., SONGER, J. G., MONTGOMERY, J. s CASSELLS, J. AS. (1986) Veterinary Record 118, 11-13 ELLIS, W. A., THIERMANN, A. B., MONTGOMERY, J., HANDSAKER, A., WINTER, P. J. & MARSHALL, R. B. (1988) Research in Veterinary Science 44, 375-379 JARVIS, B. D. W., SCOTT, K. F., HUGHES, J. E., DJORD~ JEVIe, N., ROLFE, B: G. & SHINE, J. (1983) Canadian Journal of Microbiology 29,200-209 JOHNSON, R. D. & HARRIS, V. G. (1967) Journal ofBacteriology 94,27-31 KAFATOS, F. C., JONES, C. W. & EFSTRADIADIS, A. (1979) Nucleic Acid Research 7, 1541-1552 LE FEBVRE, R. B. & THIERMANN, A. B. (1986) American Journal of Veterinary Research 47, 959-963 LE FEBVRE, R. B., THIERMANN, A. B. & FOLEY, J. (1987) Journal of Clinical Microbiology 25, 2094-2097
.
197
MANIATIS, T., FRITSCH, E. F. & SA~IBROOK, J. (1982) Molecular Cloning: a Laboratory Manual. New York, Cold Spring Harbor Laboratory MARSHALL, R. B., MANKTELOW, B. W., RYAN, T. J. & HATHAWAY, S. C. (1976) New Zealand Medical Journal 84, 74-75 MARSHALL, R. B., WILTON, B. E. & ROBINSON, A. J. (1981) Journal of Medical Microbiology 14, 163-166 MARSHALL, R. B., WINTER, P. J., THIERMANN, A. B. & ELLIS, W. A. (1985) Veterinary Record 117,669-670 ROBINSON, A. J., RAMADASS, P., LEE, A. & MARSHALL, R. B. (1982) Journal of Medical Microbiology 15,331-338 SELDIN, L. & DUBNAU, D. (1985) International Journal of Systematic Bacteriology 35, 151-154 TAYLOR, J. M., ILLMENSEE, R. & SUMMERS, J. (1976) Biochimica et Biophysica Acta 442,324-330 THIERMANN, A. B., HANDSAKER, A. L., FOLEY, J. W., WHITE, F. H. & KINGSCOTE, B. F. (1986) American Journal of Veterinary Research 47, 61-66 VAN EYS, G. J. J. M., ZAAL, J., SCHOONE, G. J. & TERPSTRA, W. J. (1988) Journal of General Microbiology 134, 567-574 YASUDA, P. H., STEIGERWALT, A. G., SULZER, K. R., KAUFMANN, A. F., ROGERS, F. & BRENNER, D. J. (1987) International Journal of Systematic Bacteriology 37,407-415 ZUERNER, R. & BOLIN, C. (1988) American Society for Microbiology D66, 861
Received Apri/2/, /989 Accepted February /9, /990