pomona leptospiral vaccine

FEMS MicrobiologyImmunology64 (1990)111-118 Publishedby Elsevier

111

FEMSIM00109

Antigens recognized by the human immune response to vaccination with a bivalent hardjo/pomona leptospiral vaccine A.J. Chapman, S. Faine and B. Adler Departmentof Microbiolo~, Monash University,Clayton, Victoria,Australia Received16 February1990 Revisionre~ivedand accepted23 April1990 Key words: Leptospira interrogans; Antibody response; Human vaccination

I. SUMMARY Serum from volunteer subjects vaccinated with a bivalent whole cell vaccine of Leptospira interrogans serovar hardjo/serovar pomona grown in protein-free medium, was tested by the micrc~ scopic asglutination test (MAT), enzyme-immunoassay (EIA) and immunoblotting. Specific IgM antibodies to either serovars hardjo or pomona were detected in some subjects as early as 6 days after vaccination with peak antibody levels occurring 13-68 days after vaccination. Whereas all subjects produced specific IgM to both serovars, not all produced specific IgG to both serovars. Immunoblottin8with hardjo sonicate revealed that all subjects produced IgM antibodies reacting with the 15, 23 and 28 kDa components of hardjo iipopolysaccharide (LPS), and most produced IgM antibodies that reacted with the 34.5 kDa flageUar doublet. In contrast, not all sera immunoblotted asainst pomona sonicate reacted with the 29 and 35 kDa components of pomona LPS. However all subjects produced antibodies reacting with a diffuse 14.4-27 kDa band. These antibodies ap-

Con~spmdenc¢to: A$. Chapman, Departmentof Mkrobiolosy, MonashUniversity,Clayton,3168Victoria,Australia.

peared early in the immune response. Serum from the one vaccinated subject tested protected hamsters from acute lethal infection with serovar pomona.

2. INTRODUCTION Leptospirosis is an acute febrile illness, the severity of which varies from mild to rapidly fatal. The causative bacterium Leptospira interrogans is classified serologically into more that 200 serovars [1]. In Australia and New Zealand, leptospirosis is most frequently caused by serovar hardjo or sero= vat pomona [2,3] usually in dairy farm workers, slanghtermen or meat inspectors [1,3-5]. Leptospirosis may remain undiagnosed because the symptoms are non-specific [6], especially in the early stages of infection [1]. Even mild illness due to leptuspirosis can result in significant economic loss, due to absence from work, and other complications such as congenital infection leading to stillbirth have been reported [7]. Currently, there is no effective, non-toxic,. leptospiral vaccine licensed for human use in Australasia, Europe or North America. Vaccines of killed leptospiral suspensions grown in media containing animal serum protect but often result

0920-85M190/$03.50© 1990Federationof EuropeanMicrobiologicalSocieties

112

in both local and severe systemic reactions [1]. More recent whole cell vaccines produced in chentically-defined protein-free medium [8] have reduced but not eliminated these reactions. However, little is known about their relative efficacy and they are not used 8euerally. Immunity to leptospirosis is thought to depend solely on the B-cell mediated production of agglutinating and opsonic antibodies [9,11-13], with resistance to re-infection appearing to depend solely on serovar- or serogroup-reaedve antigens. Second or subsequent infections, when diagnosed, usually occur with a different serovar [14]. At present, the antigens involved in natural immunity to leptospiral infection are not well understood. However we have recently reported [15] on the antigens recognized by the human immune response to infection with serovar hardjo. Convalescent serum samples reacted in immunoblotring with the major 28 kDa sub-unit of hardjo lipopolysaccharde (LPS), which appeared to be a major surface-exposed antigen, as agglutinating monoclonal antibodies also reacted with the LPS component. Other agglutinating monoclonal antibodies have shown that LPS is also a major surface component of serovar pomona [16]. However, it is not known whether vaccination is capable of stimulating the production of similar antibodies. The aim of this research was to investigate the humoral immune response, of human volunteers immunized with a killed, bivalent hardjo/pomona vaccine, derived from leptospires grown in a protein-free medium, and to determine whether the antibodies elicited by the vaccine reacted with the same antigens as those elicited by natural leptospiral infection.

land for use in serological tests. These isolates were grown in EM/H medium and prepared as previously described [15].

3.2. Preparation and administration of the hardjo / pomona vaccine Serovars hardjo and pomona which had been subcultured at least 30 times in protein-free medium were harvested by centrifugation, washed three times in pyrogen-free saline, killed with 1% (v/v) neutral buffered formalin for 1 h at 21°C, and then washed 5 times and resuspended in sterile pyrogen-frce saline. It was sterility tested (Commonwealth Serum Laboratories, Melbourne) before use. No viable leptospires were detected by culturing in EMJH medium. Six healthy adult male volunteers with no previous history of leptospirosis were vaccinated intradermally in the flexor aspect of the forearm with 0.1 ml of vaccine containing 2 × 10 s leptospires of each serovar. Blood was collected by veuepuncture before and at intervals after immunization. Full blood biochemical and haematological examination was performed on days 0, 6 and 13 by the diagnostic laboratories of the Alfred Hospital, Prahran, Melbourne. The trial conformed to ethical standards of informed consent, supervision and institutional approval.

3.3. Serological tests

3. MATERIAL AND METHODS

The microscopic a881utination test (MAT) and enzyme-immunoas~y (EIA) were performed as previously described [15], with the exception that the background optical density (OD) for the EIA was determined by averaging the EIA OD of 8 serum samples, negative by MAT, obtained from volunteers who had no previous known leptospiral infection. The passive hamster protection test was described previously [20].

3.1. Lepwspires

3.4 Preparation of lepwspiral antigens

The reference strain Hardjoprajitno of L, interrogans serovar hardjo [17] and a human isolate of I.. interrogans serovar pomona [18] were adapted

Ultrasonicated leptospiral antigen preparations were prepared and standardized for use in EIA and immunoblotting as previously described [10,15].

to growth in a protein-free medium [19] and used in the vaccine. More recently isolated field strains of serovars pomona and hardjo were obtained from R.B. Marshall, Massey University, New Zea-

3.5. Gel electrophoresisand inommoblotting

Sodium dodecyl sulphate-polyacryl~de gel

113 electrophoresis ( S D S - P A G E ) and immunoblotring were performed as previously described [15].

tions macroscopically resembled those seen in tuberculin tests. There was no systemic disturbance observed and biochemical and haematological parameters remained normal.

4. RESULTS

4.1. Reactogenicity of the vaccine preparation All volunteers experienced eryth~ma of 1-4 cm diameter and slight inflammation at the injection site for up to 72 h after injection, accompanied by some itchiness but no pain. In addition, two subjects developed central necrotic indurations, resembling small microabscesses, similar to the polymorphonuclear infiltrations previously reported in guinea-pigs and rabbits [19]. The reac-

4.2. Antibody response as measured by M A T and EIA Serum samples obtained over a 6-month period from six volunteer subjects were tested by M A T and EIA for the presence of specific anti-leptospiral antibodies. Fig. 1 shows the time course of antibody appearance in subjects 1 - 3 (results for subjects 4 - 6 not shown). Peak M A T and EIA antibody levels for all volunteers were generally observed 13-33 days after immunization. In most

HARDJO

I,. i

li

1.

SUBJECT2

U~

,o o

~

~0 O _J

Ix. I.

°.411!

\

g-o

SUBJECT3

I

11

SUBJECT3

-"12

~

~0~o,o,,o,oo,~o

-I'o

Fig. L T'ma¢course of appearance of antibodies against /.. interrogans~rovars har~o and pomona produced by subjec~ 1-3, as measured by MAT and EIA. Serumsamples were numbered accordingto the number of days after vaccination. EIA optical densities were subtracted for background. • Ik MAT; • •, EIA IgM; • 0, EIA lgG.

114 MAT titres gradually declined after the peak, although very low levels of agglutinating antibody were still detectable 182 days after vaceination. While serum from subjects 1, 2 and 4-6 contained specific IgM and IgG (as measured by EIA) for serovars pomona and hardjo, serum from subject 3 did not contain IsG antibodies reacting with serovar hardjo and only a small amount of IsG that reacted with serovar pomona. The peak levels and persistence of antibody produced against either serovar varied with each volunteer, but there was little difference in the time of antibody peak against either serovar pomona or hardjo produced by each individual subject. Antibody levels detected by EIA usually mirrored those detected by MAT, with the appearance of agglutinatin8 antibodies coinciding with EIA-detectable IsM. The appearance of IgG varied between subjects. IsM levels declined to background levels by day 182 in all subjects, with the excepticn of subject 2 in whom MAT- and EIA-detectable antibody levels increased between days 68 and 182. The appearance of specific IgM and IsG antibodies not only varied between volunteers, but also between antibody produced by the same volunteer and directed at either serovar pomona or serovar hardjo. This was most obvious in comparins the EIA profiles of subjects 1 and 2 with that of subject 3 (Fig. 1). Both subjects 1 and 2 produced IsM and IsG antibodies that reacted with serovars hardjo and pomona, but the IsG response to hardjo antigens occurred later and was of a lower magnitude than the IsM response. The IsG response of subjects 1 and 2 to serovar pomona differed in that it was equal in magnitude and occurred at the same time as the IsM response. In contrast, subject 3 produced only IsM antibodies that reacted with serovar hardjo, while the 18(3 response directed against serovar pomona was of a lower magnitude and occurred later than the 18M response. cases,

tAW

M G

M

G

M

G

M G

M G

M G

33

68

182

(Kd)

94," 67," 43,"

20.1" 14.~

O

6

13

.A

DAYS AFTER IMMUNISATION

Fig. 2. Immunoblotsof L interrogans scrovar hardjo with sequentialsenunsamplesfromsubject1. M and G denotethe detectionof IgM and lgG antibodiesrespectively.Serumsampies were numberedaccordingto the numberof days after vaccination.Numberson the left indicatethe molecularmass (kDa) of the standardproteinsuscd. with the 15,23 and 28 kDa components of hardjo LPS, and appeared between days 6 and 13 after immunization. This antibody was predominantly IsM, although some IS(} antibodies were observed reacting with the major 28 kDa component of LPS. Reaction with other hardjo antigens varied, but all volunteers produced antibody against the 34.5-35 kDa flagella doublet and most produced some antibody against the 32 kDa major outer membrane protein. Subject 1 alone produced IgG antibodies that reacted with an evenly spaced pattern of bands ranged between approximately 67-100 kDa. 18(3

tAW (Kd) 94b 67)

M G !

M

G

M G

M G

13

33

M

G

M G

43b 30~ 20.1) 14.4b 0

6

68

182

DAYS AFTER IMMUNISATION

4.3. Characterization of antibody response to serovar hardjo by immunoblotting When tested by immunoblottin8 with serovar hardjo sonicate (results for subjects 1-3 shown in Figs. 2-4) antibody from all volunteers reacted

Fig, 3. Immunoblotsof L, interrogans serovar bardjo with sequentialserumsamplesfromsubject2. M and G denotethe detectionof IgM and 18(3antibodiesrespectively.Serumsampies were numberedaccordingto the numberof days after vaccination. N u m b e r s on the left indicatethe molecular mass (kDa) of the standardproteinsused.

115 M

6

M

6

M

6

M

6

M

G

M

M

6

MW Oc~ 94P 67,

S

OR G

M

6

Ill

G

M

G

g

G

~d 94~ 671, i 43,.

43,

, .,

+ 30, 20"1~ 20-1, 14.4P

i 0

0

6 DAYS

13

33

68

182

AFTER IMMUNISATION

Fig. 4. Immtmoblots of L, interrogans serovar hardjo with sequential serum samples from subject 3. M and G denote the detection of lgM and lgG antibodies respectively. Serum samplea were nmmbegedaau3fding to the number of days after vac~mtion. Nombers on the left indicate the molecular mass (klDa)of the standard proteins used. antibodies produced by volunteers 2 and 6 also reacted with a 44.5-45 k D a doublet and the 34.535 k D a flagella doublet. However these antibodies were present in the pre-immune serum and may constitute antibodies from a previous infection, or reaction with antibodies produced against other antigens.

4.4. Characterization of antibody response to serovar p o m ~ a by immunoblotting In contrast to the reaction with serovar hardjo antisens, not all subjects produced antibodies against serovar pomona LPS. Subjects 1 a n d 2 (Fibs. 5 a n d 6) produced antibody to b o t h the 29 a n d 35 k D a components of serovar pomona LPS by day 33. However n o antibody from subject 3 (Fi 8. 7) was observed to react with serovar pomona LPS until day 182. Similarly sera from subjects 4 - 6 also had barely discernible or no reaction with serovar pomona LPS (data not shown). However all subjects produced 18M and 18(3 antibodies that reacted with a diffuse 14.4-27 kDa b a n d that appeared as early as day 6 after immunization. This diffuse b a n d appeared to begin just below the lower 29 k D a component of pomona LPS. 4.5. Passive protection of hamsters Volumes of 0.2 ml of sera from subject 2 were tested for their capacity to protect pairs of ham-

6

13

33

68

182

DAYS AFTER IMMUNISATION

Fig. 5. lmmtmoblots of L. interrogans serovar pomona with sequential serum samples from subject 1. M and G denote the detection of IgM and 18(3 antibodies respectively. Serum sampies were numbered according to the number of days after vaccination. Numbers on the left indicate the molecular mass (kDa) of the standard proteins used.

sters from acute infection with serovar pomona. Sera from days 0, 6 and 13 failed to protect but protective capacity was seen in sera from days 33, 68 and 182, which completely protected pairs of hamsters. All 4 control hamsters died within 4 days, from acute leptospirosis confirmed at autopsy. Passive M A T titres in these protected hatnsters were approximately 20 prior to challenge. It was not possible to test other sera due to geographical constraints; hamsters are not available in Australia. Similarly, it was not possible to test the ability of the serovar hardjo component to

M

G

M

G

M

G

M

G

M

G

M

G

94" 67" 43"

i

30" 20"1~ 14"4"

6

6

13

33

68

182

DAYS AFTER IMMUNISATION

Fig. 6. Immunoblots of L interrogans serovar pomona with sequential serum samples from subject 2. M and G denote the detection of IgM and IgG antibodies respectively.Serum sampies were numbered according to the number of days after vaccination. Numbers on the left indicate the molecular mass (kDa) of the standard proteins used.

116 II

S

•

S

6

Ill

S

13

II

S

33

II

II

68

III

S

182

DAYS AFTER IMMUNISATION

Fig. 7. immonoblotsof L interrogans ~rovar pomona with sequentialserumsamplesfrom subject3. M and O denotethe detectionof IBMand lgG antibodiesrespectively.Serumsampies were numbered according to the number of days after vaccination.Numbers on the left indicatethe molecularmass (kDa) of the standardproteinsm,ed. stimulate protective antibodi,.~ as serovar hardjo does not cause acute infections in hamsters. 5. DISCUSSION Although a lack of correlation has been repo_~d between MAT titre and EIA OD for individual sera [10,15,22,23], our results agreed with our previous fmdings [15] that the times of appearance of antibodies measured by MAT and EIA in patients' sera were similar and exhibited similar profiles during the course of the immune response. The differences observed between the IgM and IgG EIA profiles (Fig. 1) of the subjects may indicate a secondary response in subjects 1 and 2 where the IgM and 18(3 peaks occurred simultaneously. For example, subject 2 participated in a leptospiral vaccine trial seven years earlier, which had involved the administration of whole cells of the saprophytic L. &'flexa serovar patoc. Although this did not apparently affect the agslutinating antibody response to pomona or hardjo in the current trial, it is possible that the EIA detected an ananmestie response to cross-reactive non-agslutinating antigens [25]. Serovars pato¢,/um~o and pomona do not share detectable agglutinating antigens [25]. Previously we found that during natural leptospiral infection, spedfic IgM levels detected by EIA declined after a strong response, while specific IgG levels remained high [15]. Comparing the EIA 18M responses of subjects 1-6, all, except

subject 2 showed this decline to background levels over the 182 day sampling period. Although the IgG response varied between subjects, it was noted that where high IgG levels were initially found these levels remained higher than the IgM levels. Subject 2 was the only person whose antibody levels increased between days 68 and 182. Immunoblotting sera from subjects 1-6 with har~o sonicate revealed that all produced 18M antibodies against the 15, 23 and 28 kDa components of hardjo LPS described previously [24]. Reaction with other hardjo antigens varied, although reactiun of IgM a n d / o r IgG antibodies with the 34.5-35 kDa flagella doublet was observed with serum from most subjects. This was similar to our previous study on the human humoral immune response to naturally acquired infection with serovar hardjo [15], that LPS is the major antigen involved in the immune response to serovar hardjo. lmmunoblottin$ sera from subjects 1-6 with pomona sonicate showed that, in contrast to the results observed with hardjo sonicate, not all subjects produced antibody that reacted with pomona LPS. However, sera from subjects that had the higher MAT titres to serovar pomona always reacted with serovar pomona LPS on immunoblots. When antibodies reacted wi:h serovar pomona LPS, it was difficult to discern the reactivity to other pomona antigens, such as the flagella, as the two diffuse LPS subunits covered a region rangin8 from approximately 27-40 kDa [16]. It is also sisnificant that the vaocinees" sera protected hamsters from acute infection, similar to the protection reported for sera from naturally a~luired infections [20], in which anti-LPS ag81urinating antibodies were involved. Although it was not possible to test anti-har~o antibodies for protective capacity it is reasonable to assume that they would also protect, in view of the correlation between agglutinatin& opsonizin8 and protective antibodies [13,24,25]. Our results thus confn-m that immunization of humans with protein-frec grown vaec~ne can induce the production of protective anti-LPS antibodies which appear to have a central role in immunity to infection, similar to the response following natural infection and to protection in animals [13,15,20].

117 The major serovar pomona ant]sea identified b y sara from all subjects was a diffuse 14.4-27 k D a band. This b a n d was similar in size a n d shape to the species-spocific a n d genus-specific antigen that reacted with h u m a n anti-hardjo serum [15] a n d hyperimmune rabbit anti-leptospiral antismurn [25] respectively, This antigen does not appear to b e involved in a881utination as day 6 serum from subject 3 reacted with this b a n d in immunoblots, b u t h a d n o M A T titre. There was n o difference seen between the immunoblot profiles of sera from subject 3 o n days 6 a n d 13, yet the day 13 serum h a d the maximum M A T titre produced b y that subject. It is possible that in addition to LPS there m y b e hitherto unidentified protein a881utinating antigens present on the leptospiral membrane, which were not detected due to their denaturation during the immunoblotring process. This has been previously observed in Escherichia col], where monocional antibodies reacted only with the native, trimeric, outer memb r a n e protein [26]. The nature of this low molecular mass antigen is unknown, but warrants further study as it is clearly a major antigen reengnised by b o t h infected a n d vaccinated humans. Furthermore, as this antigen appears to be cross-reactive a n d to elicit antibodies early in the immune response to leptmpiral infection, it may b e useful as the basis of a defined cross-reactive diagnostic test for h u m a n leptospirosis. Undefined eross-reactive preparations have previously been used for this purpose [10,21,27~81.

ACKNOWLEDGEMENTS This work was supported b y a research grant from the National Health and Medical Research Council, Canberra, Australia, W e thank Dr. W J . Spicer and staff at Alfred Hospital for assistance with monitoring vaccine administration and clinica] laboratory tests.

[2] Adler, B. and Faine, S. (1980) F.pidemio~ of human leptospirosis in AustralL~ Comm. Dis. Int. (Aust. Dept Health) S0/21, 2-5. [3] Blackmore, D.K. and SchoHum, L. (1982) Risks of con. tract]rig leptospirosis on the dairy farm. N.Z. Med. J. 95, 649-652.

[4] Mackintosh, C.G., Schollum, L.M., Harris, R.E., Blackmore, D.K., Willis. A.F, Cook, N.R. and Stoke, LC.R. (19s0) Epidcmiology of kpu~piro~s in dairy farm workers in the Manawatu. L A cross-sectional serological survey

and associated occupational factors. N.Z. Vet. J. 28, 245250. [5] Swart, KS., Wilks, C.R., Jackson, K.B. and Hayman, J.A. (1983) Human leptospirosis in Victoria. Mcd. J. Aust. 1, 460-463. [6] Kauffmann, A.F. (1976) F.pidemiolos~ trends of lepto* spirosis in the United States, 1965-1974. in The Biolo~ of Parasitic Spirochetes.(Johnson, R.C., Ed.), pp. IT?-189. Academic Press, London. [7] Faine, S., Adler, B., Christopher, W. and Valentine, R. (1984) Fatal congenital human leptospirosis. Zblt. Bakt. Hyg. A 257, 548. [81 Shenber& E. and Torten, M. (1973) A new leptospiral vaccine. !. Development of a vaccine from leptospires grown in a chemicallydefined medium. J. Infect. Dis. 128, 642-646. [9] Adler, B. and Faine, S. (1977) Host immunologicalmechanisms in the resistance of mice to leptospiral infections. Infect. lnunon. 17, 67-72. [10] Adler, B., Murphy, A.M., Locamini, S.A. and Faine, S. (1980) Detection of specific anti-leptmpiral immonoglobulins M and G in human serum by solid-phase ew zym~linked immuno~fbant a,~ay. J. Clin. MicrobinL11, 452-457. 01] Adler, B., Faine, S., Muner, H.K. and Green, D.F. (1980) Maturation of the humoral immune response determines the susceptibility of guinea pip to leptospirmis. Patholosy 12, 529-538. [12] McGrath, H.E., Adler, B., Vinh, T. and Faine, S. (1984) P h a l ~ o s i s of virulent and avimlant leptospires by pis and human pobnmqCaonudear leudgoojtesin vitro. Pathology 16, 243-249. [13] Jost, B.H., Adler, B., Vinh, T. and Faine, S. (1986) A monoclonal antibody rcec~ S with a determinant on leptcepirai lipopolysaceh~de worsts guinea piss , ~ i , l t leptmpirmis. J. Med. Microbiol. 22, 269-275. ]14] Doherty, R.L. (1956) A second infection with leptospimsis. Med. J. Aust. 1, 59-60. [15] Chapman, AJ., Adler, B. and Faine, S. (1988) Antigens by the human immune response to infection with Leptospira interroga~ serovar hardjo. J. Mod. ~/Ugrobiol. 25, 269-278.

REFERENCES [1] Faine, S. (Ed.) (1982) Guidelines for the Control of Leptospirmh. Offset Publ. No. 67, Wld. Hlth. Or&, Oeneva.

[16] Adler, B., Ballard, S.A., Miller, S.J. and Fai~, S. (1989) Monodonal antibodies reacting with serogroup and serovat spccific epitop~ on differ~t lipopolysaccheridesubunits of Lepwspira interrogam serovar pomona. FEMS Microbiol. Immunol. 47, 213-218.

118 [17] Wolff, J.W. (1953) The Classification of Pathogeific Leptospires. Symposium ou the Leptospires. Med. Sci. Publ. No. 1, US Govt. Print. Off., Washington, D.C., 174-185. [18l Christmas, B.W., Bra88er, J.M. and Till, D.G. (1974) Dairy farm fever in New Zealand. Isolation of L. pomona and I, hardjo from a local outbreak. N.Z. Med. J. 79, 904-906. [19] Christopher, W.L., Adler, B. and Faine, S. (1982) lmmunogenicity of ieptospiral vaccines grown in protein.free medium. J. Med. Microbiol. 15, 493-450. [20] Adler, B. and Faine, S. (1978) The antibodies involved in the human immune response to leptospiral infection. J. Med. Microbiol. 11, 387-400. [21] Terpstra, WJ., Ligthart, G.S. and Schoone, G.J. (1980) Serodiagnosis of human leptospirosis by enzyme-linkedimmunosorbent-assay (ELISA). Zblt. Bakt. Hyg. A. 247, 400-405. [22] Miiner, A.R., Jackson, K.B., Woodruff, K. and Smart, IJ. (1985) Enzyme-linked immunosorbent assay for determinins specific immonoglobulin M in infections caused by Lepta~pira interrogans serovar hardjo. J. Clin. Microbiol. 22, 539-542.

[23] Farrelly, H.E., Adler, B. and Faine, S. (1987) Opsonic monoclonal antibodies against lipopolysaeeharide antigcns of Leptospira interrogans serovar hardjo. J. Med. MierobioL 23,1-7. [24] Adler, B. and Faine, S. (1976) Susceptibility of mica treated with cyclopbosphamide to Icthal infection with Leptospira interrogans scrovar pomona. Infect. Immun. 14, 703-708. [25] Chapman, AJ., Adler, B. and Faine, S. (1987) Genusspecific antisens in Icptospiza revealed by immunoblotring. Zblt. BakL Hyg. A. 264, 279-293. [26] Pages, J.-M., Pases, C., Bernadac, A. and Prince, P. (1988) Immtmolosical evidence for differences in the exposed re~ons of OmpF porins from F~cherichia coil B and K-12. Mol. lmmunol. 25, 555-563. [27] Cox, C.D., Alexander, A.D. and Murphy, L.C. (1957) Evaluation of the hemolytic test in the serodlagnosis of human I o p t o s p ~ J. Infect. Dis. 101, 210-218. [28] Palit, A. and Gulasekharam (1973) Genus-specific leptospiral antigen and its possible use in laboratory diagnosis. J. Ciin. Pathol. 26, 7-16.