Deoxyribonucleic Acid Relatedness Between Serogroups and Serovars in the Family Leptospiraceae with Proposals for Seven New Leptospira Species PAUL0 H

Home , Leptospira, Leptospiraceae

INTERNATIONAL JOURNAL OF SYSTEMATIC BACTERIOLOGY, OCt. 1987, p. 407415 Vol. 37, No. 4 0020-7713/87/040407-09$02.00/0 Copyright 0 1987, International Union of Microbiological Societies

Deoxyribonucleic Acid Relatedness between Serogroups and Serovars in the Family Leptospiraceae with Proposals for Seven New Leptospira Species PAUL0 H. YASUDA,' ARNOLD G. STEIGERWALT,* KATHERINE R. SULZER,3 ARNOLD F. KAUFMANN,4* FAYE ROGERS,3 AND DON J. BRENNER2 Department of Microbiology, Instituto de Ciencias Biomedicas da Universidade de Sao Paulo, Sao Paulo, Brazil 05508, and Molecular Biology Laboratory,2 Leptospirosis Lab~ratory,~and Bacterial Zoonoses A~tivity,~Division of Bacterial Diseases, Center for Infectious Diseases, Centers for Disease Control, Atlanta, Georgia 30333

Deoxyribonucleic acid hybridization (hydroxyapatite method, 55 and 70°C) was used to characterize 38 serovars from 22 named serogroups of Leptospira interrogans and Leptospira bijlexa, 6 serovars from 4 new unnamed serogroups of Leptospira interrogans, and single serovars of the proposed species Leptospira parva and Leptonema illini. Deoxyribonucleic acid relatedness confirmed the validity of Leptospira parva and Leptonema illini. The well-accepted species Leptospira interrogans and Leptospira bijlexa, as currently defined, were extremely heterogeneous. Relatedness results revealed at least five new species among the parasitic serovars formerly included in Leptospira interrogans and two new species among the saprophytic serovars formerly included in Leptospira bijlexa. Serogrouping did not equate with species identification, as serovars from several different subserogroups belonged to different species. The new species named in this paper are Leptospira, noguchii, Leptospira weilii, Leptospira santarosai, Leptospira borgpetersenii, Leptospira meyeri, Leptospira wolbachii, and Leptospira inadai.

Leptospiraceae is a family in the Spirochaetales whose Similar to salmonellae, the serovars (serotypes) of lepto- type genus is Leptospira (7, 12, 14). The three Leptospira spires are named, and antigenically related serovars are species are Leptospira interrogans, which contains a large organized into serogroups. Leptospira parva contains a number of serogroups whose strains are parasitic or patho- single serovar (13), and Leptonema illini contains two genic for humans and animals (14); Leptospira bijlexa, which serovars (1, 12). At least 19 serogroups that contain more contains a large number of serogroups whose strains are than 170 serovars are known for Leptospira interrogans, and primarily found in fresh surface waters and moist soil and are 38 serogroups with 65 serovars have been described for the rarely isolated from humans or animals (14); and Leptospira less-studied species Leptospira bijlexa (14). parva, which was isolated from tap water, is not pathogenic In previous DNA hybridization studies (2, lo), six DNA for hamsters, and is biochemically intermediate between L. relatedness groups were identified from 15 serovars. Among interrogans and L. bijlexa (13). Leptonema, with the single the pathogenic serovars, Leptospira interrogans serovars species Leptonema illini, was proposed as a second genus in australis, bataviae, icterohaemorrhagiae, kabura, muen- the Leptospiraceae (12). Leptonema illini deoxyribonucleic chen, and pomona were in one relatedness group, and acid (DNA) has a guanine plus cytosine (G+C) content of 51 serovars javanica, tarassovi, and perhaps celledoni were in to 53 mol% (12), compared with 47.4 mol% for Leptospira a second relatedness group. Among the saprophytic sero- parva (13), 36 to 39 mol% for Leptospira bijlexa (14), and 35 vars, Leptospira bijlexa serovars andamana, patoc, and to 41 mol% for Leptospira interrogans (14). Leptonema saopaulo were in one relatedness group, serovar codice was possesses cytoplasmic tubules which are absent in in a second, serovar ranarum was in a third, and Leptonema Leptospira, and the structure of the basal complex on its illini was in a fourth. flagella resembles that of gram-positive bacteria, whereas These studies indicated that leptospires contain a great this structure in Leptospira resembles that of gram-negative deal of heterogeneity. In this study we determined the DNA bacteria (12). Leptonema strains were isolated from the relatedness of a much larger sample of leptospires in an urine of a healthy bull, a turtle, and water. They are not attempt to improve the classification of this complex and pathogenic for animals and are biochemically similar to important group of bacteria. Leptospira bijexa except for their ability to grow in Trypticase soy broth (1, 12). In addition to pathogenicity, Leptospira interrogans differs from Leptospira bijlexa by its MATERIALS AND METHODS inability to grow in the presence of 8-azaguanine and its inability to grow at 13°C (14-17). Two biogroups within Bacterial strains. A total of 45 leptospire serovars from the Leptospira interrogans differ in their ability to grow in the three previously named species of Leptospira and 1 serovar presence of 2,6-diaminopurine (14). Leptospira parva does of Leptonema were studied. These organisms represent 18 not grow in the presence of either of the above chemicals and named and 4 unnamed serogroups of Leptospira inter- does grow at 13°C (13). In general, the phenotypic charac- rogans, 4 serogroups of Leptospira bijlexa, Leptospira teristics that reportedly differentiate the leptospires have parva, and Leptonema illini. Some of the serogroups are been based on evaluations of a limited number of strains. subdivided into subserogroups based on antigenic relatedness patterns for convenience in serotyping. Serovars representing these subserogroups were included in the study. * Corresponding author. Only the reference type strains for the serovars in the study

407 408 YASUDA ET AL. INT. J. SYST.BACTERIOL.

TABLE 1. Strains of Leptospiraceae used in DNA relatedness studies Serovar Serogroup Subserogroup Strain Species assignment icterohaemorrhagiae Icterohaemorrhagiae Icterohaemorrhagiae RGAT (= ATCC 43642T) Lep tospira in terrogans copenhageni M20 Leptospira interrogans smith i Smithi Smith Leptospira interrogans sarmin Sarmin Sarmin Leptosp ira weilii canicola Canicola Canicola Hond Utrecht IV Lep tospira in t e rrog a n s schueflneri Schueffneri Vleermuis 90 Leptospira interrogans ballum Ballum Mus 127 Leptospiva borgpetersenii javanica Javanica Veldrat Batavia 46T (= ATCC 43292T) Leptospira borgpetersenii celledoni Celledoni CelledoniT (= ATCC 43285T) Lebtosbiru weiiii pomona Pomona Pomona Leptospira interrogans australis Aus tralis Australis Ballico Leptospira interrogans jalna Jalna Jalna Leptospira interrogans autumnalis Autumnalis Autumnalis Akiyami A Leptospira interrogans fortbragg Fortbragg Fort Bragg Lep tospira noguc hii louisiana Louisiana LSU 1945 Leptospira noguchii pyrogenes Pyrogenes Pyrogenes Salinem Leptospira interrogans zanoni Zanoni Zanoni Leptospira interrogans grippotyphosa Grippot yphosa Moskva V Lep tospira in t errogans djasiman Djasiman Djasiman Leptospira interrogans cynopteri Cynopteri 3522 C Pending study“ panama Panama CZ 214KT (= ATCC 43288T) Lept ospira nog uchii hebdomadis Hebdomadis Hebdomadis Hebdomadis Leptospira interrogans borincana Borincana HS 622 Leptospira santarosai mini Mini Sari Leptospira borgpetersenii sejroe Sejroe Sejroe M 84 Leptospira borgpetersenii WO& Wolffi 3705 Leptospira interrogans saxkoebing Saxkoebing Mus 24 Leptospira interrogans bataviae Bataviae Van Tienen Leptospira interrogans tarassovi Tarassovi Tarassovi Perepelicin Leptospira borgpetersenii bakeri Bakeri LT 79 Leptospira santarosai atlantae Atlantae LT 81 Leptospira santarosai navet Navet TRVL 109873 Leptospira santarosai shermani Shermani LT 821T (= ATCC 43286T) Leptospira santarosai peru New unnamed 1 MWlO Leptospira santarosai banana1 Aa14 Leptospira santarosai wuwain MW6 Leptospira santarosai “AGC” New unnamed 2 AGC Leptospira interrogans “84-011370” New unnamed 3 84-011370 Leptospira noguchii lyme New unnamed 4 loT (= ATCC 43289T) Leptospiva inadai ranarum Ranarum Iowa City FrogT (= ATCC 43287T) Leptospira meyeri codice Codice CDCT (= ATCC 43284T) Leptospira wolbachii semaranga Semaranga Veldrat Semarang 173 Leptospira meyeri patoc Patoc IT (= ATCC 23582T) Leptospira bifexa andaman Andamana CH 11 Lep tospira bif exa parva NA~ HT (= NCTC 11395T) Leptospiru parva illini NA 3055T (= ATCC 43287T) Leptonema illini ‘ Study of more serovars is needed for reliable species identification on the basis of DNA relatedness. NA, Not applicable.

were used (Table 1). In the remainder of this paper we refer DNA methods. A11 strains were grown at 30°C in polysor- to leptospires by serovar name only (e.g., smithi instead of bate albumin medium (23). Cells were harvested by centrif- Leptospira interrogans serovar smithi) unless specifically ugation during the late logarithmic or stationary phase of noted otherwise. The term leptospires refers to members of growth. Unlabeled DNA was isolated and purified as previ- the genus Leptospira as well as to Leptonema illini. Strains ously described (3). DNA was labeled in vitro with 32P(11). were maintained in a polysorbate albumin medium (PLM-5; DNA relatedness was determined by the hydroxyapatite Armour Pharmaceutical Co., Kankakee, Ill.). The identity of method (3), except that, because of the low G+C contents of each strain was verified by our leptospirosis laboratory (a the DNAs of strains, hybridization reactions were done at an Food and Agricultural Organization/World Health Organiza- optimal reassociation temperature of 55°C and at a stringent tion International Reference Center), using the agglutinin- incubation temperature of 70°C. G+C compositions were absorption test with specific antisera before DNA extraction determined in triplicate by thermal denaturation (18). (5). Two strains, AGC and 84-011370, were previously Phenotypic characteristics. The study strains were tested unreported isolates from humans in Peru and Panama, for their ability to grow at incubation temperatures of 11, 30, respectively; these isolates are members of two new un- and 37°C. The lowest incubation temperature of 11°C was named serovars, representing two new unnamed serogroups used due to technical limitations of the incubation system. of Leptospira interrogans. A description of serovar lyme, Although the ability to grow at 13°C is often cited as a including some information included in this report, has been differential characteristic between pathogenic and sapro- published recently (21). phytic strains, the study on which this criterion is based VOL. 37. 1987 DNA RELATEDNESS IN LEPTOSPIRES 409 found that comparable growth occurred at 10°C except for (subserogroup Mini), Sejroe (subserogroup Sejroe), and serovar andaman, which had a slower growth response than Tarassovi (subserogroup Tarassovi). Serovar cynopteri other saprophytic serovars when the initial inoculum titer (serogroup Cynopteri) was the only serovar in its relatedness was low (15). Each strain was concurrently tested in dupli- group, as was serovar lyme (new serogroup 4). cate at each incubation temperature, using polysorbate albu- Of the five nonpathogenic Leptospira serovars, andaman min medium. Growth was measured by visual inspection for (serogroup Andamana) and patoc (serogroup Semaranga) turbidity on days 0, 7, 14, 21, and 28. Viability of strains were in one DNA relatedness group; ranarum (serogroup without apparent growth was measured by subculture into Ranarum) and semaranga (serogroup Semaranga) were in a polysorbate albumin medium on the same days. The cultures second relatedness group; and codice (serogroup Codice) that were incubated at the optimal growth temperature of was in a third relatedness group (Table 2). The single 30°C were considered internal controls. serovars representing the species Leptospira parva and Growth responses to 8-azaguanine (225 pg/ml) and 2,6- Leptonema illini were in relatedness groups separate from diaminopurine (10 pg/ml) were tested by using essentially each other and from all of the DNA relatedness groups listed the method of Johnson et al. (15, 17). Each strain was tested above (Table 2). Leptospira parva was 0 to 3% related to in duplicate at an incubation temperature of 30"C, using other leptospires, and Leptonema illini was 0 to 14% related polysorbate albumin medium. Growth was measured by to other leptospires. The three relatedness groups containing visual inspection for turbidity on days 0, 7, 14, 21, and 28. nonpathogenic serovars were 0 to 49% interrelated and were Viability of strains without apparent growth was measured 1 to 35% related to the relatedness groups containing patho- by subculture into polysorbate albumin medium without the genic serovars (Table 2). respective purine analog on the same days. In addition, The average levels of relatedness for pathogenic serovars concurrent control cultures without an added purine analog from any one relatedness group were 84 to 99% in 55°C were incubated and evaluated for growth in the same man- reactions and 74 to 96% in 70°C reactions (Table 3). The level ner. of divergence in related sequences was 0.5 to 2.5%. Between Growth response to copper sulfate was tested by the groups of pathogenic serovars, with the exception of method of Fuzi and Csoka (8). Each strain was tested in Leptospira inadai serovar lyme, the average levels of relat- duplicate at an incubation temperature of 30°C by using edness ranged from 26 to 75% in 55°C reactions and from 7 polysorbate albumin medium containing varying concentra- to 56% in 70°C reactions, with a level of divergence of 5.0 to tions of copper sulfate (0, 25,50, 100, 1,000, and 10,000 ppm 15.5% in related sequences (Table 3). [0, 25, 50, 100, 1,000, and 10,000 pglml]). Growth was G+C determinations were done for the DNA from the measured by visual inspection for turbidity on days 0, 7, 14, type strain of each species (Table 4). The pathogenic type 21, and 28. Viability of strains without apparent growth was strains, from Leptospira interrogans, the four newly desig- measured by subculture into polysorbate albumin medium nated species formerly included in Leptospira interrogans, without added copper sulfate on the same days. and Leptospira interrogans serovar cynopteri, had G +C The lipase (trioleinase) activity of the strains was deter- values of 35.2 mol to 40.7 mol%. The nonpathogenic type mined by the method of Johnson and Harris (15). strains, from Leptospira bijlexa and the two newly desig- nated species formerly included in Leptospira b@exa, had RESULTS G+C values of 33.5 mol to 38.8 mol%. The G+C contents of Leptospira parva and Leptonema illini were 53.1 and 54.2 DNA results. A total of 46 serovars representing 18 of the mol%, respectively, and the G+C content of Leptospira 19 recognized pathogenic serogroups, 4 of the 38 nonpatho- inadai (serovar lyme) was 42.6 mol%. genic serogroups, and 6 other serogroups (4 previously Phenotypic characteristics. Varying numbers of serovars unreported pathogenic serogroups, Leptospira parva, and belonging to all of the DNA relatedness groups showed Leptonema illini) were tested for DNA relatedness. The limited growth up to 21 days of incubation at 11°C; however, pathogenic strains formed six DNA relatedness groups (Ta- none of the serovars showed growth at 28 days. Despite the ble 2). The icterohaemorrhagiae relatedness group con- lack of growth, all serovars remained viable throughout the tained 18 serovars from Leptospira serogroups Ictero- incubation period at this temperature. Only the serovars of haemorrhagiae (subserogroups Icterohaemorrhagiae and Leptospira inadai, Leptospira bijlexa, Leptospira meyeri, Smithi), Australis (subserogroups Australis and Jalna), Leptospira parva, and Leptonema illini grew at 37°C for 28 Autumnalis (subserogroup Autumnalis), Bataviae, Canicola days (Table 5). All other serovars died at this temperature (subserogroups Canicola and Schueffneri), Djasiman, Grip- before 28 days. All serovars grew at 30°C. potyphosa, Hebdomadis (subserogroup Hebdomadis), With the exception of andaman, all of the saprophytic Pomona, Pyrogenes (subserogroups Pyrogenes and Zanoni), serovars grew in the presence of 8-azaguanine (Table 5). The and Sejroe (subserogroups Saxkoebing and Wolffi) and strain growth of all of the pathogenic strains except Leptospira AGC of new serogroup 2. panama was the reference serovar inadai serovar lyme was inhibited by 8-azaguanine. With the for a relatedness group that included serogroups Autumnalis exception of Leptospira bijlexa serovar andaman, Lepto- (subserogroup Fortbragg), Panama, and Pyrogenes (sub- nema illini, and Leptospira parva, the saprophytic serovars serogroup Louisiana) and strain 84-011370 of new serogroup grew in the presence of 2,6-diaminopurine. Of the 39 patho- 3. celledoni was the reference serovar for a relatedness genic serovars, 15 also grew in the presence of this purine group with serovars from serogroups Celledoni and analog. Lipase activity was observed in 20 of the 39 patho- Icterohaemorrhagiae (subserogroup Sarmin). The shermani genic serovars and all 7 saprophytic serovars. Lipase activ- relatedness group contained serovars from serogroups ity and the ability to grow in the presence of 2,6-diamino- Shermani , Hebdomadis (subserogroup Borincana), and purine tended to be inversely related. Of 27 lipase-positive Tarassovi (subserogroups Atlantae, Bakeri, and Navet) and serovars, only 6 grew in the presence of 2,6-diaminopurine, the three serovars (bananal, peru, and wawain) of new whereas 13 of 19 lipase-negative serovars did grow in media serogroup 1. The javanica relatedness group contained containing this purine analog. A similar relationship was not serovars from serogroups Javanica, Ballum, Hebdomadis observed with 8-azaguanine, another purine analog. 410 YASUDA ET AL. INT. J. SYST.BACTERIOL.

TABLE 2. DNA relatedrless of leptospire serovars“ Labeled DNA from serovar: icterohaemorrhagiae copenhageni smithi panama celledoni RBR RBR RBR RBR RBR RBR RBR RBR Unlabeled DNA from RBR at D (%) at at at D (%) at at D (%) at serovar: at D(%)“ at 55”Cb 70°C 55°C 70°C 55°C 55°C 70°C 55°C 70°C Icterohaemorrhagiae 100 0.0 100 99 0.0 100 100 75 8.0 47 41 14.0 14 copenhageni 92 0.0 100 100 0.0 100 75 8.5 52 45 smithi 10 100 12 11 canicola 93 1.0 93 70 schueffneri 100 100 pomona 100 1.5 92 74 47 australis 100 100 jalna 100 100 autumnalis 100 100 pyrogenes 100 100 zanoni 100 90 grippotyphosa 100 0.0 99 djasiman 100 98 saxkoebing 100 1.0 70 43 bataviae 100 95 “AGC” 100 0.0 100 72 8.0 51 wolfi 98 0.0 72 46 hebdomadis 89 3 .O 71 52 12.5 24 panama 77 8.0 50 74 6.5 45 73 100 0.0 100 50 13.5 16 fortbragg 74 72 5.5 41 86 2.5 77 louisiana 74 5.5 70 6.5 45 84 2.0 77 “84-011370” 82 8.0 81 3.5 68 celledoni 38 11.5 8 31 48 13.5 16 100 0.0 100 sarmin 42 11.5 12 52 88 2.0 77 shermani 39 12.5 48 19 46 41 13.0 56 10.0 borincana 47 55 9 52 61 8.5 42 bakeri 50 13 47 16 58 9.5 35 atlantae 39 10.0 12 46 14 57 10.0 32 navet 62 8.0 45 11.5 20 51 13.5 32 banana1 52 11 50 21 54 8.5 46 Peru 44 9.5 17 54 29 62 8.5 42 wawain 52 11 50 22 60 9.0 42 javanica 53 16 45 11.0 13 38 60 10.5 30 68 7 .O 54 ballum 57 15 75 18 sejroe 42 9.0 15 mini 41 37 tarassovi 55 15 lyme 5 6 15.0 3 7 7 cynopteri 65 53 58 5.0 44 63 64 6.5 51 47 11.5 31 andaman 27 31 11 21 35 19 14 patoc 5 6 6 ranarum 2 9 7 2 4 semaranga 3 15.5 6 7 codice 9 4 5 5 parva 0 0 1 3 illini 7 3 6 14

Four of the seven saprophytic serovars but only one and in which related DNAs contain 6% or fewer unpaired pathogenic serovar (Leptospira inadai serovar lyrne) were bases (divergence) (3). Exceptions to one of these three able to grow in media containing 100 ppm of copper sulfate parameters are sometimes encountered, but rarely does a (Table 5). All serovars grew in media Containing 1,000 and serovar not conform to at least two parameters, including 10,000 ppm of copper sulfate. percent divergence. In this study the DNA hybridization groups exemplified by Leptospira interrogans serovar icterohaemorrhagiae and Leptospira noguchii serovar pan- DISCUSSION ama were 70 to 75% interrelated in 55°C reactions; however, the level of relatedness fell to 44 to 45% in 70°C reactions, We have defined a species as a group of serovars whose and the level of divergence was 6.5 to 9.0%. In contrast, the DNAs are 70% or more related at an optimal reassociation members of the Leptospiru interrogans relatedness group temperature (55°Cin this study) and are 55% or more related were 99% related at 55°C and 96% related at 70°C and at a stringent reassociation temperature at which only showed 0.5% divergence; and the members of the Lepto- closely related sequences can hybridize (70°C in this study) spira noguchii relatedness group were 84% related at 55°C VOL. 37, 1987 DNA RELATEDNESS IN LEPTOSPIRES 411

TABLE 2-Conrinued

shermani javanica andaman ranaritm codice --parva lyme RBR RBR RBR RBR RBR RBR RBR RBR RBR RBR RBR RBR at D (%) at at D (%) at at D (%) at at D (%) at at D (%) at at at 55°C 70°C 55°C 70°C 55°C 70°C 55°C 70°C 55°C 70°C 55°C 70'C 39 12.5 16 30 13.5 4 2 5 0 3 48 33 14.5 7 3 14 2 2 28 5

30 3

23 3

28 29 21 14.0 41 13.0 23 36 13.5 11 2 0 5 32 13.5 33 14.0 29 56 10.0 46 53 8.5 43 5 4 0 7 66 7.5 50 100 0.0 100 62 6.5 5 1 0 5 90 0.5 91 55 10.5 29 92 2.0 88 51 12.0 29 93 0.5 87 52 10.5 27 86 2.0 86 51 22 86 0.5 90 57 10.0 36 86 0.5 91 56 10.0 29 86 0.0 86 62 9.0 32 62 6.5 56 100 0.0 100 6 1 0 94 0.0 99 92 4.5 90 97 80 2 88 2.5 83 5 6 3 1 0 100 34 15.5 15 26 13.5 3 1 0 0 5 23 15 100 0.0 100 6 0 4 6 3 90 4.0 86 23 1 1 21 100 0.0 100 23 0 2 8 3 83 2.5 86 49 11.5 14 0 5 6 2 21 25 100 0.0 100 1 7 3 2 0 0 100 72 7 4 0 3 2 2 6

a '*P04-labeled DNAs were reacted with unlabeled DNAs from the same strain (homologous reaction) and from a series of other leptospires. Each reaction was done at least twice. The level of reassociation in homologous reactions ranged from 31 to 82% (average, 57%) before normalization. Control reactions in which labeled DNA was incubated in the absence of udlabeled DNA showed 0 to 40% (average, 1.1%)binding to hydroxyapatite. These control values were subtracted before normalization. ' Relative binding ratio (RBR) = [(percentage of DNA bound to hydroxyapatite in heterologous reactions)/(percentage of DNA bound in homologous reactions)] x 100. ' Calculations of percent divergence (D) assumed that a 1% decrease in the thermal stability of a heterologous DNA duplex compared with that of the homologous duplex was caused by 1% of the bases within the duplex that were unpaired. Final values were rounded to the nearest 0.5%. and 74% related at 70°C and showed 2.5% divergence. The strains will have to be examined in order to determine data clearly indicate that these relatedness groups represent whether cynopteri represents a new species. separate species. Previous DNA relatedness studies of leptospires were Serovar cynopteri was 65% or less related to labeled done by Haapala et al. in 1969 (10) and by Brendle et al. in DNAs from other leptospires. Labeled cynopteri DNA was 1974 (2). Twelve of the serovars used in our study were used 90% related to icterohaemorrhagiae and panama in 55°C in one or both of the previous studies. For 11 of these reactions (data not shown), with 6 to 7.5% divergence, but serovars the results were in total agreement. Haapala et al. showed only 37 to 46% relatedness in 70°C reactions. Be- placed Leptospira weilii serovar celledoni in the same group cause these differences are small, additional leptospire with Leptospira borgpetersenii serovars javanica and taras- 412 YASUDA ET AL. INT. J. SYST.BACTERIOL.

TABLE 3. Summary of DNA relatedness data for parasitic leptospires" Labeled DNA from relatedness group represented by serovar: Unlabeled DNA from icterohaemorrhagiaed panama celledoni shermani javanica relatedness group represented by serovar: RBR RBR RBR RBR RBR RBR RBR RBR RBR RBR at D (%)" at at D (%) at at D (%) at at D (%) at at D (%) at 55"Cb 70°C 55°C 70°C 55°C 70°C 55°C 70°C 55°C 70°C icterohaemorrhagiae 99 0.5 96 70 9.0 44 43 1.4 14 44 12.5 16 28 14.0 7 panama 75 6.5 45 84 2.5 74 50 13.5 16 41 13.0 23 33 13.5 celledoni 40 11.5 10 50 13.5 16 88 2.0 77 56 10.0 46 61 8.0 47 shermani 49 10.0 13 48 12.5 20 57 9.5 39 88 1.0 88 56 9.5 29 javanica 49 10.0 15 57 10.5 24 68 7.0 54 62 6.5 56 93 2.5 88 cynopteri 62 5.0 49 64 6.5 51 47 11.5 31 34 15.5 15 26 13.5 Relatedness and divergence data were taken from Table 2 and are presented as arithmetic means obtained from the total number of serovars tested. The results of homologous control reactions (100% by definition) were not included in the calculations. Relatedness groups are referred to by the name of the serovar that was labeled or, if not labeled, by the only serovar in the group (serovars cynopteri and illini). See Table 2, footnote 6. See Table 2, footnote c. Includes reactions, in which both icterohaemorrhagiae and copenhageni were used. sovi. Serovar javanica was more than 95% related to taras- less than the value previously reported, and Leptospira sovi, with 0.2% divergence, and was 78% related to meyeri serovar ranarum, whose G+C content is 7.7 mol% celledoni, with 6.4% divergence. In our study, Leptospira lower than the value previously reported. In the previous borgpetersenii serovars javanica and tarassovi were 88% report, however, the G+C contents were determined by the related, with 2.5% divergence; Leptospira weilii serovars buoyant density method and not by thermal denaturation. celledoni and javanica were 68 and 53% related, with 7.0 and Until recently, essentially all pathogenic serovars were 8.5% divergence, respectively, in reciprocal reactions. placed in Leptospira interrogans, and saprophytic (usually Therefore, there is really no disagreement between our data nonpathogenic) serovars were placed in Leptospira bifexa. and those of Haapala et al. Hovind-Hougen et al. proposed Leptospira parva as a third The G+C contents of type serovars which we found Leptospira species (13), and Hovind-Hougen also proposed (Table 4) are generally in agreement with those of previous a second genus, Leptonema, with a single species, Lepto- studies (2,10,12,13). Two exceptions are Leptospira bifexa nema illini (12). These species have not yet gained general serovar patoc, whose G+C content (36.0 mol%) is 3 mol% acceptance (14). Our data support the validity of both species; Leptospira parva was 0 to 3% related to other leptospires, and Leptonema illini was 0 to 14% related to other leptospires. TABLE 4. G+C contents of leptospires Leptospira interrogans is the type species of Leptospira. G+C content (mol%) Its type strain is Leptospira interrogans serovar ictero- Strain haemorrhagiae RGA (= ATCC 43642). Therefore, all sero- This study !::!:: vars in the same relatedness group as Leptospira interrogans serovar icterohaemorrhagiae are Leptospira interrogans Leptospira interrogans serovar ictero- 34.9 * 0.9 35.5 haemorrhagiae strain RGAT serovars (Tables 1 and 2). Although cynopteri may represent Leptospira noguchii serovar panama 36.5 2 1.2 a separate species, it is retained in Leptospira interrogans strain CZ 214KT pending further study. All pathogenic serovars not tested in Leptospira weilii serovar celledoni strain 40.5 * 0.7 38.3 (Bd)b this study also must remain in the species Leptospira inter- CelledoniT rogans pending further study. About one-halfof the serovars Leptospira santarosai serovar shermani 40.7 * 0.6 tested from serogroups (or subserogroups) formerly included strain LT 821T in the species Leptospira interrogans were not in the Leptospira borgpetersenii serovar 39.8 * 0.3 40.4 Leptospira interrogans DNA relatedness group. We placed javanica strain Veldrat Batavia 4BT these serovars in the following five new species: Leptospira Leptospira meyeri serovar ranarum 33.5 * 0.2 41.2 (Bd) strain Iowa City FrogT borgpetersenii, Leptospira inadai, Leptospira noguchii, Leptospira meyeri serovar semaranga 35.2 * 0.2 Leptospira santarosai, and Leptospira weilii (Tables 1and 2) strain Veldrat Sernarang 173 (see species descriptions below). The type serovars for the Leptospira wolbachii serovar codice 37.2 t 0.5 38.0 (Bd) proposed species are the serovars whose DNAs were se- strain CDCT lected in a dropout fashion for reference comparison. Leptospira bifiexa serovaf patoc strain 36.0 5 0.3 39.0 Five saprophytic serovars, all previously classified in the Patoc IT species Leptospira bifexa, formed three DNA relatedness Leptospira biflexa serovar andaman 37.6 5 0.7 39.1 groups. The type strain, Leptospira bifexa serovar patoc strain CH 11 strain Patoc I (= ATCC 23582), and serovar andaman Leptospira inadai serovar lyme strain 42.6 2 0.9 loT remain in the species Leptospira bifexa. All serovars from Leptospira parva serovar parva strain 48.7 * 0.4 47.4 (Bd) saprophytic serogroups not tested in this study also remain HT in the species Leptospira bgexa pending further study. The Leptonema illini serovar illini strain 54.2 2 0.2 53.0 (Bd) four saprophytic serovars that were not related to Lepto- 3055* spira bifexa at the species level were placed in two new species , Lep tospira m ey eri and Leptospira wol bac hii (Ta- a These values were taken from references 2, 10, 12, and 13. Bd, Determined by the buoyant density method. If not indicated other- bles 1 and 2) (see species descriptions below). wise, G+C values were determined by the thermal denaturation method. The phenotypic characteristics of the species will need VOL. 37, 1987 DNA RELATEDNESS IN LEPTOSPIRES 413

TABLE 5. Phenotypic characteristics of leptospires used in DNA relatedness studies No. which grow at an No. which grow in the No. of incubation temp of: presence of: No. having Species strains lipase 11°C 30°C 8-Aza- 2,6-Di-amino- Copper activity 37"c guanine purine sulfate Leptospira interrogans 19 0 19 0 0 4 0 16 Leptospira borgpeterseni 5 0 5 0 0 4 0 0 Leptospira inadai 1 0 1 1 1 0 1 1 Leptospira noguchii 4 0 4 0 0 0 0 3 Leptospira santarosai 8 0 8 0 0 7 0 0 Leptospira weilii 2 0 2 0 0 0 0 0 Lep rospira biflexa' 2 0 2 1 1 1 1 2 L ep tospira mey eri' 2 0 2 2 2 2 2 2 Leptospira wolbachii" 1 0 1 0 1 1 0 1 Leptospira parva" 1 0 1 1 1 0 1 1 Leptonema illini" 1 0 1 1 1 0 0 1 Species composed of saprophytic strains.

further definition. Even under the classification scheme positive. Most do not grow in Trypticase soy broth and do extant at the present time (14), such definition is needed. not require NaCl for growth. However, some saprophytic Characteristics proposed as differentiating features between serovars isolated from seawater are halophilic. The DNA Leptospira spp. are few and have been based on reactions of relatedness relationships of the serovars are shown in Table a limited number of Leptospira strains. Studies of these 2. Other characteristics of the seven new species are given characteristics will be more meaningful when DNA related- below and in the tables and text. ness studies have been completed on a larger proportion of Leptospira noguchii sp. nov. Leptospira noguchii (no. gu' the defined serovars. Such studies should include, to the chi.i. N.L. gen. n. noguchii to honor Hideyo Noguchi, the extent feasible, multiple strains of each serovar. At this time, Japanese microbiologist who named the genus Leptospira) overinterpretation of these data should be avoided. does not grow at 11 or 37"C, and its growth is inhibited by Previously, leptospires were classified by morphology, a 8-azaguanine (225 pg/ml), 2,6-diaminopurine (10 pg/ml), and few biochemical reactions, pathogenicity, and serology. copper sulfate (100 ppm). Lipase is usually produced. It Pathogenicity and serology, although important, have contains serovars from serogroups Panama, Autumnalis proved to be generally poor criteria on which to place great (subserogroup Fortbragg), and Pyrogenes (subserogroup weight for taxonomic purposes in many genera, including Louisiana) and from a new unnamed serogroup represented Salmonella, Clostridium, Klebsiella, Yersinia, and Vibrio. In by strain 84-011370. The type strain, strain CZ 214 (= ATCC the case of the Leptospira, our study has shown that the 43288), is a member of serogroup Panama and serovar pathogenic strains are contained in at least six separate panama and was isolated by N. B. Gale et al. from the species. Similarly, saprophytic strains from only 4 of the 38 kidney of an opossum (Didelphis marsupialis) trapped in recognized serogroups were in three separate species. Sero- Panama (9); it has a G+C content of 36.5 k 1.2 mol%. logical classification also did not correlate well with species Leptospira weilii sp. nov. Leptospira weiiii (weil' i. i. N.L. identification. Among the saprophytic serovars, serovars gen. n. weilii to honor Adolph Weil, the German physician semaranga and patoc are both in serogroup Semaranga, but who was among the first to clinically differentiate lepto- are in different species (Leptospira meyeri and Leptospira spirosis [Weil's disease] from other types of infectious bifiexa, respectively). Among pathogenic serovars, serovars jaundice) does not grow at 11 or 37"C, and its growth is that belong to different subserogroups (5) in serogroups inhibited by 8-azaguanine (225 pg/ml), 2,6-diaminopurine (10 Autumnalis, Hebdomadis, Icterohaemorrhagiae, Pyrogenes, pg/ml), and copper sulfate (100 ppm). Lipase is not pro- Sejroe, and Tarassovi are in different species. duced. It contains serovars from serogroups Celledoni and This study provides a significant advance in the classifi- Icterohaemorrhagiae (subserogroup Sarmin). The type cation of the Leptospiraceae. We have confirmed the exist- strain, strain Celledoni (= ATCC 43285), is a member of ence of and partially determined the boundaries of the four serogroup Celledoni and serovar celledoni and was isolated previously described species and have proposed (see below) from the blood of a patient in North Queensland, Australia, seven new species on the basis of DNA relatedness. A by D. J. W. Smith et al. (22); it has a G+C content of 40.5 possible 12th species, represented by serovar cynopteri, * 0.7 mol%. requires additional study. Further work is necessary to Leptospira santarosai sp. nov. Leptospira santarosai (san. determine the proper classification of serovars that were not ta. ros' ai. N.L. gen. n. santarosai to honor Carlos A. Santa tested, especially among the saprophytic strains. This infor- Rosa, the Brazilian veterinary microbiologist who was a mation will form a rational framework on which to base pioneer in the study of leptospirosis as a human and animal studies of the phenotypic characteristics of the Leptospiru health problem in Brazil) does not grow at 11 or 37°C. Its species. growth is inhibited by copper sulfate (100 ppm) and 8- Taxonomic proposals. Nomenclatural proposals are given azaguanine (225 pg/ml), but growth is usually not inhibited below for seven new species that conform to the definition of by 2,6-diaminopurine (10 pg/ml). Lipase is not produced. It the family Leptospiraceae and to the definition of the genus contains serovars from serogroups Shermani, He bdomadis Leptospiru. The cells are gram negative, flexible, and heli- (subserogroup Borincana), and Tarassovi (subserogroups cal. These organisms are motile by means of two flagella Atlantae, Bakeri, and Navet) and serovars bananal, peru, (axial fibriles). They are obligately aerobic and oxidase and wawain from a new unnamed serogroup. The type 414 YASUDA ET AL. INT.J. SYST.BACTERIOL. strain, strain LT 821 (= ATCC 43286), is a member of LITERATURE CITED serogroup Shermani and serovar shermani and was isolated from the kidney of a spiny rat (Proechimys semispinosus) 1. Bazovska, S., K. Hovind-Hougen, A. Rudiova, and E. Kmety. 1983. Leptospira sp. strain Dimbovitza, first isolate in Europe trapped in the Panama Canal Zone by N. B. Gale (24); it has with characteristics of the proposed genus Leptonema. Int. J. a G+C content of 40.7 k 0.6 mol%. Syst. Bacteriol. 33:325-328. Leptospira borgpetersenii sp . no v . L ep tospira b o rgp e tersen ii 2. Brendle, J. J., M. Rogul, and A. D. Alexander. 1974. Deoxyri- (borg. pe' ter. sen. i. i N.L. gen. n. borgpetersenii to honor bonucleic acid hybridization among selected leptospiral sero- C. Borg-Petersen, the Danish physician who made signifi- types. Int. J. Syst. Bacteriol. 24:205-214. cant early contributions to the epidemiology and microbiol- 3. Brenner, D. J., A. C. McWhorter, J. K. Leete-Knutson, and ogy of leptospirosis in Europe) does not grow at 11 or 37°C. A. G. Steigerwalt. 1982. Escherichia vulneris: a new species of Its growth is inhibited by 8-azaguanine (225 pg/ml) and Enterobacteriaceae associated with human wounds. J. Clin. copper sulfate (100 ppm), but growth is usually not inhibited Microbiol. 15: 1133-1 140. 4. Diesch, S. L., W. F. McCulloch, J. L. Braun, and H. C. by 2,6-diaminopurine (10 pg/ml). Lipase is not produced. It Ellinghausen, Jr. 1966. Leptospires isolated from frog kidneys. contains serovars from serogroups Javanica, Ballum, Nature (London) 209:939-940. Hebdomadis (subserogroup Mini), Sejroe (subserogroup 5. Dikken, H., and E. Kmety. 1978. Serological typing methods of Sejroe), and Tarassovi (subserogroup Tarassovi). The type leptospires, p. 259-307. In T. Bergan and J. R. Norris (ed.), strain, strain Veldrat Bataviae 46 (= ATCC 43292), is a Methods in microbiology, vol. 11. Academic Press, Inc., New member of serogroup Javanica and serovar javanica, was York. isolated from a Java house rat (Rattus brevicaudatus) in 6. Esseveld, H., and A. Mochtar. 1938. Over het voorkomen van Indonesia by M. Sardjito et al., and was first serotyped by H. een nieuw leptospira-type (L.javanica) en het 'Salinem-type' bij Esseveld (6, 20); it has a G+C of 39.8 k 0.3 mol%. veldratten op Java. Geneeskd. Tijdschr. Ned.-Ind. 78: Leptospira meyeri sp. nov. Leptospira meyeri (mey' er. i. 1513-1522. 7. Faine, S., and N. D. Stallman. 1982. Amended descriptions of N.L. gen. n. meyeri to honor Karl F. Meyer, the veterinarian the genus Leptospira Noguchi 1917 and the species L. inter- who established veterinary public health in the United States rogans (Stimson 1907) Wenyon 1926 and L. bifIexa (Wolbach through his broad interests in the zoonoses, including and Binger 1914) Noguchi 1918. Int. J. Syst. Bacteriol. leptospirosis) does not grow at 11°C but does grow at 37°C. 32:461463. Growth occurs in the presence of either 8-azaguanine (225 8. Fuzi, M., and R. Csoka. 1960. Die Diffenenzierung de Fg/ml), 2,6-diaminopurine (10 pg/ml), or copper sulfate (100 pathogenen und saprophytischen Leptospiren mittels eines ppm). Lipase is produced. It contains serovars from Kupfersulfattestes. Zentralbl. Bakteriol. Parasitenkd. Infek- serogroups Ranarum and Semaranga (serovar semaranga). tionskr. Hyg. Abt. 1 Orig. 179:231-237. The type strain, strain Iowa City Frog (= ATCC 43287) is a 9. Gale, N. B., A. D. Alexander, L. B. Evans, R. H. Yager, and R. G. Metheney. 1966. An outbreak of leptospirosis among U.S. member of serogroup Ranarum and serovar ranarum and Army troops in the Canal Zone. 11. Isolation and characteriza- was isolated from the kidneys of a leopard frog (Rana tion of the isolates. Am. J. Trop. Med. Hyg. 1564-70. pipiens) in the United States by S. L. Diesch et al. (4); it has 10. Haapala, D. K., M. Rogul, L. B. Evans, and A. D. Alexander. a G+C content of 33.5 5 0.2 mol%. 1969. Deoxyribonucleic acid base composition and homology Leptospira wolbachii sp. nov. Leptospira wolbachii (wol. studies of Leptospira. J. Bacteriol. 98:421428. bach' i. i. N.L. gen. n. wolbachii to honor S. B. Wolbach, 11. Hickman-Brenner,F. W., G. P. Huntley-Carter, Y. Saitoh, A. G. the American microbiologist who first identified Leptospira Steigerwalt, J. J. Farmer 111, and D. J. Brenner. 1984. [Spirochaeta] biflexa) does not grow at either 11 or 37°C. Moellerella wisconsensis, a new genus and species of Entero- Growth occurs in the presence of either 8-azaguanine (225 bacteriaceae found in human stool specimens. J. Clin. Micro- pg/ml) or 2,6-diaminopurine (10 pg/ml). Growth is inhibited biol. 19:46O-463. 12. Hovind-Hougen, K. 1979. Leptospiraceae, a new family to by copper sulfate (100 ppm). Lipase is produced. It contains include Leptospira Noguchi 1917 and Leptonema gen. nov. Int. a serovar from serogroup Codice. The type strain, strain J. Syst. Bacteriol. 29:245-251. CDC (= ATCC 43284), is a member of serogroup Codice and 13. Hovind-Hougen, K., W. A. Ellis, and A. Birch-Andersen. 1981. serovar codice and was isolated from water in the United Leptospira parva sp. nov.: some morphological and biological States by M. K. Ward (19); it has a G+C content of 37.2 5 characters. Zentralbl. Bakteriol. Parasitenkd. Infektionskr. 0.5 mol%. Hyg. Abt. 1 Orig. Reihe A 250:343-354. Leptospira inadai sp. nov. Leptospira inadai (in. ad' ai. 14. Johnson, R. C., and S. Faine. 1984. Family 11. Leptospiraceae, N.L. gen. n. inadai to honor Ryokichi Inada, the Japanese p. 62-67. In N. R. Krieg and J. G. Holt (ed.), Bergey's manual microbiologist who is regarded by some to have first isolated of systematic bacteriology, vol. 1. The Williams & Wilkins Co., leptospires from human patients) does not grow at 11°C but Baltimore. 15. Johnson, R. C., and V. G. Harris. 1967. Differentiation of does grow at 37°C. Growth occurs in the presence of either pathogenic and saprophytic leptospires. I. Growth at low tem- 8-azaguanine (225 pg/ml) or copper sulfate (100 ppm), but peratures. J. Bacteriol. 94:27-31. does not occur in the presence of 2,6-diaminopurine (10 16. Johnson, R. C., and V. G. Harris. 1968. Purine analogue Fg/ml). Lipase is produced. The type strain, strain 10 (= sensitivity and lipase activity of leptospires. Appl. Microbiol. ATCC 43289), is a member of serogroup Lyme and serovar 16:1584-1590. lyme; it was isolated from the skin of a patient who had 17. Johnson, R. C., and P. Rogers. 1964. Differentiation of patho- concurrent, but unrelated, illness due to Lyme disease by G. genic and saprophytic leptospires with 8-azaguanine. J. Bacte- P. Schmid et al. (21) and has a G+C content of 42.6 2 0.9 rial. 88:1618-1623. mol%. 18. Marmur, J., and P. Doty. 1962. Determination of the base composition of deoxyribonucleic acid from its thermal denaturation temperature. J. Mol. Biol. 5109-118. ACKNOWLEDGMENT 19. Nardelli, M. G., and B. Babudieri. 1972. Studio sistematico di 20 ceppi di leptospire acquicole. Ann. 1st. Super. Sanita 8:114- P.H.Y. was supported by grant 83/1766-0 from Fellowship 121. FAPESP (Fundagao de Amparo a Pesquisa do Estado de Sao 20. Sardjito, M., A. Mochtar, and R. Wirasmo. 1937. Het voorko- Paulo). men van leptospiren bij veldratten (Rattus rattus brevicaudatus) VOL. 37, 1987 DNA RELATEDNESS IN LEPTOSPIRES 415

in verband met het eventueele besmettingsgevaar bij het baden V. M. MacDonald, C. J. Ross, and R. L. Doherty. 1954. The in de Rawa Pening. Meded. Dienst. Volksgezondheid Ned.-Ind. serological classification of 89 strains of leptospirae from North 26:24-39. Queensland. Australas. Ann. Med. 3:98-105. 21. Schmid, G. P., A. C. Steere, A. N. Kornblatt, A. F. Kaufmann, 23. Sulzer, C. R., and W. L. Jones. 1980. Leptospirosis, methods in C. W. Moss, R. C. Johnson, K. Hovind-Hougen, and D. J. laboratory diagnosis, revised ed. Centers for Disease Control Brenner. 1986. Newly recognized Leptospira species (“Lep- publication 80-8275. Centers for Disease Control, Atlanta. tospira inadai” serovar lyme) isolated from human skin. J. Clin. 24. Sulzer, K. R., V. Pope, and F. Rogers. 1982. New leptospiral Microbiol. 24:484-486. serotypes (serovars) from the Western Hemisphere isolated 22. Smith, D. J. W., H. E. Brown, J. I. Tonge, C. N. Sinnamon, during 1964 through 1970. Rev. Latinoam. Microbiol. 24:15-17.