Biological activities of Leptospiral Lipopolysaccharide

Zbl. Bakt. Hyg. A 261, 53-64 (1986)

Biological Activities of Leptospiral Lipopolysaccharide EMIKO ISOGAIt, HIROSHI ISOGAI 2 , YOICHI KUREBA YASHI 3 , and NOBUYOSHI ITO l 1 2

3

Department of Preventive Dentistry, Department of Oral Anatomy, School of Dentistry, Higashi Nippon Gakuen University, Ishikari Tobetsu 061-02, Central Laboratory, Daiichi Seiyaku, 1-16-13 Kitakasai, Edogawa-ku, Tokyo 134, Japan

With 8 Figures· Received May 15, 1985 . Accepted August 26, 1985

Summary Lipopolysaccharide extracted with phenol-water from Leptospira interrogans serovar copenhageni strain Shibaura (L-LPS) showed various biological activities. In lethality for mice, L-LPS was active (LD 50, 3.4 mg/mouse) but about 12 times less potent than Escherichia coli LPS (E-LPS) per weight basis. L-LPS had pyrogenicity for rabbits, and the fever curves showed no evidence of the classical biphasic fever produced by E-LPS. In the bone marrow of mice, L-LPS caused hemorrhages and necrosis but less severe than those caused by E-LPS. Histopathologically, fresh hemorrhages were found in the intestine, spleen, lung and the other organs at 24 h after inoculation of L-LPS. Necrosis was also found in these organs and was particularly severe in mice inoculated with more than 2 mg LLPS. Liver necrosis was found at 7th day after inoculation of L-LPS but not after inoculation of E-LPS. L-LPS had adjuvant activity just like E-LPS. L-LPS enhanced non-specific resistance to Salmonella infection and activated mouse peritoneal macrophages to kill these organisms. L-LPS was positive in limulus test just like E-LPS. These results demonstrated similarities of L-LPS and E-LPS. Some toxic effects of L-LPS were less than those of E-LPS, but some effects of L-LPS were more than those of E-LPS. L-LPS was antigenically active and the specificity was serogroup-associated, L-LPS was composed of carbohydrate (54 %), lipid (12 %), protein (5 %). Arabinose, xylose andrhamnose were major sugars as detected by gas chromatography. 2-keto-deoxyoctanate (KDO) was not detectable.

Zusammenfassung Durch Phenolwasserextraktion aus Leptospiren der Serovar copenhageni Stamm Shibaura gewonnene Lipopolysaccharide (L-LPS) zeigten verschiedene biologische Aktiviraten. Sie waren letal fur Mause (LDso, 3, 4 mg/Maus) aber etwa 12 mal weniger wirksam als Escherichia coli Lipopolysaccharide (E-LPS) bezogen auf die Dosis. L-LPS wirkte im Kaninchen pyrogen. Die erzeugten Fieberkurven hatten aber nicht den typischen zweiphasigen Verlauf, den E-LPS erzeugt, Histopathologisch wurden frische Hamorrhagien im Darm, Milz, Lunge und anderen Organen 48 Std. nach der Gabe von L-LPS gefunden. Lebernekro-

54

E.lsogai, H.lsogai, Y. Kurebayashi, and N.lto

sen entstanden 7 Tage nach der Inokulation von L-LPS, aber nicht nach der Inokulation von E-LPS. L-LPS steigerte die unspezifische Resistenz gegen Salmonella-Infektionen und aktivierte Mauseperitonealmakrophagen zur Abtotung dieser Keime. L-LPS war im LimulusTest ebenso positiv wie E-LPS. Diese Ergebnisse zeigten also Ahnlichkeiten zwischen L-LPS und E-LPS. Einige toxische Wirkungen der L-LPS waren schwacher als diejenigen des E-LPS und andererseits waren einige Effekte der L-LPS starker als die von E-LPS. L-LPS war antigenwirksam und die Spezifitat der Antigene ist mit den Serogruppen assoziiert. L-LPS besteht aus 54 % Kohlenhydraten, 12 % Lipiden und 5 % Protein. Arabinose, Xylose und Rhamnose waren die hauptsachlichen Zucker, die sich mit der Gaschromatographie feststellen lief~en. 2-keto-deoxyoctanate (KDO) wurde nicht nachgewiesen.

Introduction Leptospirosis is characterized by fever, hemorrhage and leptospiremia. For some reasons, the clinical features and histopathological findings in leptospiral infections are considered to be due to endotoxin elaborated by or released from lysed leptospiras. Firstly, similarities between the clinical and histological findings observed in this disease and those in animals given endo toxins from Gram-negative bacteria strongly favor the theory of leptospiral lipopolysaccharide as responsible for the pathogenic action of

Leptospira. Secondly, a loss of virulence for animals is not characterized by loss of hemolytic, esterase, lipase, aminopeptidase or phospholipase C activities (22). So far, no biological activity of leptospiral LPS has been reported, although several attempts have been made to demonstrate endotoxin properties. Lipopolysaccharide (F4) was not lethal for mice nor did it induce local Shwarzman reactions in rabbits (8). Arean et al. suggested that L. interrogans serovar icterohaemorrhagiae has no endotoxin or, if it had any, it could be extremely labile that it is destroyed by chemical agents used in trichloroacetic acid, phenol-water, and ethyl ether extractions (1). The aim of the present study is to determine the biological activities of leptospiral LPS, and compare them with those of LPS of E. coli in order to gain a better understanding of the role of LPS in the pathogenesis of leptospiral infection. The report describes that the leptospiral LPS has a unique composition and displays various biological potency.

Materials and Methods

Animals. Five week old ddY male mice (ddY Shizuoka strain, a random bred strain) were mainly used. Seven week old ddY, C3H, C57BL, DBA2 and ICR male mice were also used for bone marrow reactions. New Zealand white rabbits weighting 3.0-3.5 kg were used for pyrogenicity test. LPS preparation. L-LPS was extracted from Leptospira interrogans serovar copenbageni virulent strain Shibaura using the hot phenol-water technique (28). L-LPS was purified by differential centrifugation and ribonuclease treatment until spectrophotometrically free of RNA and protein. E-LPS (isolated from E. coli 0111 by phenol-water technigue, Sigma Co.) was used as control. L- and E-LPS for studies were dissolved in distilled water containing 50 ~M MgCI2 • Chemical analysis of L-LPS was done as follows. The total carbohydrate was determined by the phenol-HjaO; method (7) with glucose as standard. Hexose was deter-

Biological Activities of Leptospiral Lipopolysaccharide

55

mined by the anthrone-H 2S0 4 method (26) with glucose as standard. Pentose was estimated by the cysteine-Hjo'O, method (5) with arabinose as standard. Estimation of 6-deoxyhexose was carried out by the method of Dische and Shettles (6) with rhamnose as standard. Hexosamine was determined according to Gardell (10) with glucosamine as standard. Determination of uronic acid was performed by the carbasol-HjvO, method (2) with glucuronic acid as standard. KDO was determined by the barbituric acid method as modified by Karkhanis et al. (15). Protein was estimated according to Lowry et al. (17) with bovine albumin as standard. For analysis of neutral sugars and fatty acids, a sample of LLPS was hydrolysed for 18 h in 2N HCI at 100°C in the nonadecanoic acid and inocitol used as internal standard. The hydrolysate was partitioned between chloroform and water. The aqueous phase was used for carbohydrate analysis. The chloroform phase was used for lipid analysis. The contents were derivatized and analyzed by Gas-liquid chromatography (Shimazu Co.) using a OV-1 column (sugar analysis) and Silar 7P column (lipid analysis). Mouse lethality. Varying intravenous doses of either- or E-LPS were given to each mouse in groups of 10 animals. Deaths were recorded after injection to 7 days and 50 % lethal dose (LD50) was determined. Pyrogenicity in rabbits. The average normal rectal temperature of each rabbit was previously determined. L- or E-LPS was injected into the marginal vein of the ear. Rise in temperature from the initial value was termed [), T. Bone marrow reactions. In the present study, this test was performed by the methods of Yoshida et al. (29, 30). The bone marrow was observed by direct smear method (29) and change in number of red blood cells (RBC) per the 200 nucleated cells was examined by correlative count method (30). Limulus assay. Activity to gelate the amoebocyte lysate of.the horseshoe crab, Tachypleus tridentatus, was assayed with Pre-Gel reagent (Teikoku-zouki Co.). The activity was expressed as the lowest concentration in nanograms for 0.2 ml of Pre-Gel needed to form a solid gel. Antibody response against SRBC. Hemolytic plaque assay was done by the method of ferne et al. (14). Briefly, each mouse in groups of 20 was treated with a 10 ug dose of L- or E-LPS intravenously at same time of immunization with 2 x 10 8 sheep red blood cells (SRBC) intravenously. Spleen cell suspensions were prepared at various times after injection and plated in agar gel containing SRBC. Plaque forming cells (PFC) per 1 x 10 6 spleen cells were recorded in each group. Sera were also collected for agglutinin assay at various times after injection (24). Enhancement of non-specific resistance. Mice were intravenously inoculated with 0.25-1000 ug of L-LPS or with 0.25-25 [!g of E-LPS. After 24 h, they were challenged intraperitoneally with 0.1 ml of a 20 h culture of Salmonella typhimurium 7629, containing 2 X 10 9 cells/ml. LD50 of S. typhimurium was 3.2 x 10 6 /mouse. The number of deaths in each group consisted of 10 mice was recorded after the challenge infection. Phagocytosis in vitro. The method for phagocytosis was modified from that of Young et al. (31). The mice were inoculated with 5 % starch 48 h before injection of LPS and sacrified 24 h after intraperitoneal injection of 2.5 [!g of L-LPS or E-LPS, and the cell exudate was collected from the peritoneal cavity with Hanks balanced salt solution containing 5 U heparin and counted in a hemocytometer. In the phagocytosis test the standard mixture consisted of 0.5 ml of peritoneal macrophages (106 cells/ml, to give a final concentration of 5 5 x 10 cells/ml) and a 0.1 ml suspension of the leptospiras (to give a final concentration of 5 5 x 10 organisms/ml), with the volume of 1 ml consisting of combination of 0.1 ml normal mouse serum as a source of complement and a Hanks balanced salt solution. The mixture was incubated at 3rC for 0, 30, 60, 90 and 120 min. Samples (0.1 ml) were taken at each time, and serially diluted in cold distilled water, and plated on the DHL agar medium for counting the number of surviving organisms. Over 95 % of the macrophages were viable as determined by trypan blue exclusion. Histopathological examinations. The intestine, lung, spleen and other organs of the mice inoculated with L-LPS were examined for histopathological changes. Tissues were collected,

56

E.Isogai, H.Isogai, Y. Kurebayashi, and N. Ito

fixed in 10 % formalin, embedded in paraffin, sectioned and stained by hematoxylin and eosin. Serological examination. Microscopic agglutination test was done according to the method of WHO (32). L. interrogans serovars copenhageni (Shibaura), icterohaemorrhagiae, mankarso, naam, sarmin, ndambukie, birkini, mwogolo, boguere, ndambari, australis, autumnalis, pomona, canicola, pyrogenes were used as antigens. Anti-copenhageni Shibaura LPS antiserum was prepared in a rabbit as follows: 1 mg of LPS was emulsified in 1 ml of distilled water then mixed with 1 ml of complete Freund's adjuvant. The mixture was injected intracutaneously into 10 sites on the shaved back and into the foot (2 times at 7 day intervals). The antiserum was obtained at the 3rd postinoculation week. Immunodiffusion was done by using L-LPS (1 mg/ml, 10 Ill/well) for antigen. Immune sera against each strain were obtained as follows. The first intravenous injection of formalin-killed leptospiras (2 x 108, 1 rnl) was followed 7 days later by 7 intravenous injections of concentrated leptospiras at 5 to 7 day intervals. The concentrated leptospiras were prepared from 10-100 ml culture (approximately 2 x 108Ieptospiras/ml) by centrifugation, killed by formalin, washed with M/100 phosphate buffer and finally suspended in 1 ml of the same buffer.

Results Lethality in mice. As shown in Fig. 1, L-LPS killed inoculated mice. The LD 50 of LLPS was 3.4 mg/mouse, which was about 12 fold lower than that of E-LPS. Death of the mice inoculated with L-LPS or E-LPS occurred within 24 h. All of the mice injected with L- or E-LPS showed loss of body weight at 24 h. Pyrogenicity. L-LPS injected intravenously into rabbits at doses of 4 and 400 ug/kg caused temperature rise (Fig. 2) and so did E-LPS at a dose of 0.04 ug and 4 ug/kg. No temperature rise was seen in the 3 control rabbits and 2 rabbits inoculated with 0,04 ug L-LPS. Rabbit fever curves in response to E-LPS were biphasic, however the responses to L-LPS were clearly monophasic. Bone marrow reaction. Bone marrow of the mice inoculated with L- or E-LPS was examined at 24 h after inoculation, because the .bone marrow is one of the most

1001===0..--.--+-....- •

•06

• 12 .25

.5

1

LPS (mg/mouse)

2

4

Fig. 1. LPS-induced lethality in mice. Symbols, e, L-LPSj 0, E-LPS.

8

Biological Activities of Leptospiral Lipopolysaccharide

57

3

2

o 2

345

6

HOURS AFTER INOCULATION

Fig. 2. Pyrogenic response curve to LPS (Mean values of 3 rabbits). Symboles, e, 4 ~g/kg of L-LPS; 0, 4 ug/kg of E-LPS; _, 400 ug/kg of L-LPS; A, control. CIl

..... .....

0

I'sl U Q

I'sl

~

I'sl

.....

3.0

u

0

;::l

:z

0 0

0 0

•••

0

N

l-l


Po

,..... 2.5 0

OIl 0

-'

'-'

o

~

0 0

p:: ~

0

2.0

I'sl

•

:z

•

p::

~ ;::l

.015 .03

.06 .12

.25

LPS (mg/mouse)

Fig. 3. Bone marrow reaction in mice. Symbols,

.5

e, L-LPS; 0,

E-LPS.

58

E.Isogai, H.Isogai, Y. Kurebayashi, and N. Ito

sensitive organs to endotoxins and marked pathologic changes take place there in the early stage of endotoxico sis. Severe necrosis and hemorrhages were o bserved in smear preparations of the bone marrow of mice injected with either L-LPS or E-LPS but not in control mice. RBC value (log 10 RBC counts/nucleated cells) was increased in the bon e marrow of mice inoculated with L-LPS (Fig. 3). The higher the concentration of L-LPS used, the higher was the RBC value. Increase in RBC value was also seen in the mice injected with E-LPS. RBC value of L-LPS inoculated mice was low er tha n th at of mice inoculated with E-LPS. The mean RBC value in 12 cont rol mice was 1.7 ± 0.09. Pathological examinations. Upon necrop sy at 24 h, there was mild congestion or hemorrhage in the viscera of mice in jected with more than the LD 50 do se of L-LPS. In these mice, the intestinal loops were distended and dusky. Histop athologically, all the visceral organs show ed variab le degrees of congestion. Fresh hemorrhages were found in the intestines, spleen, lung s and other organs. Necrosis or degenerat ion was also seen in these organs. Lesions were severe in mice inoculated with more than 2 mg L-LPS. In mice inoculated with less than 0.5 mg L-LPS, lesions were mild. Degeneration of the epithelial cells of the intestine was ob served in the mice inocul ated with 2 mg L-LPS. Th e mucosal lesions varied considerably in severity depending upon the dose of L-LPS. In the spleen, the red pulp s wer e congested. Lymphoid follicles were rather atrophic with hemorrhages and lymph ocytes in the follicles were degenerated in mice inoculated with 2 or 4 mg L-LPS (Fig. 4A). Degeneration of lymphocytes in th e follicles was also seen in mice inoculated with 1 mg L-LPS, but activatio n of lymphocytes could be recognized. In the lungs, ther e was mild to severe hemorrhagic edema with infiltration of mononuclear cells (Fig. 4B). Fibr in thrombi were sometimes found in the vessels of vario us organs, but disseminated intravascular coa gulati on (DIC) was not ob served .

Fig. 4. Histopathological changes at 24 h after L-LPS inoculation. A , Degeneration and hemorrhage in the follicle of the spleen of the mice inoculated with 4 mg of L-LPS. B, Hemorrhagic edema with infiltration of mononuclear cells in the lung of the mice inoculated with 2 mg of L-LPS. C, Focal degeneration of the kidney of the mice inoculated with 2 mg of L-LPS.

Biological Activities of Leptospiral Lipopolysaccharide

59

-:

100 CJ:l H

CJ:l

0

PI:

U r.l

Z

PI: r.l

50

> H

...l N

o

.12

.25

.5

1

LPg (mg!mouse)

2

Fig. 5. Liver necrosis at 7th day after L-LPS injection. Symbols, e, L-LPS; 0, E-LPS. Focal degenerations were seen in the liver and kidneys (Fig. 4C). Control mice inoculated with distilled water showed no histopathological lesions. At 7th day after L-LPS injection, necropsy revealed dose-dependent severity of necrosis and enlargement of the liver in inoculated mice (Fig. 5). Histopathologically, there were focal areas of necrosis with moderate infiltration by neutrophils and macrophages. The spleen was enlarged

3 U)

...l ...l r.l U

Z r.l r.l ...l ~

U)

'"0

2

........

.-. 0

eo

.... ...... 0

;:l

~ ~ ~

0

PI: r.l

I:Q

S Z 235

7

DAYS AFTER INOCULATION

Fig. 6. Effect of LPS in antibody producing cells. Symbols, e, 10 ug of L-LPS; 0, 10 ug ofELPS; "'", control.

60

E.Isogai, H.Isogai, Y. Kurebayashi, and N. Ito

and prol iferation of lymphocytes was ob served. Weight ratio of spleen to bod y weight increased with dose in mice injected with L-LPS. There were onl y slight lesions in the kidne y and slight hemorrhage in the lung . Adjuvant effect of L-LPS on the antib ody respons e to SR BC. Administration of LLPS increased antibody form ation as did administration of E-LPS. The titer s were not onl y higher but also returned to normal more slowly than tho se of the LPS untreated control group. As show n in Fig. 6, the number of PFC in the L-LPS-treated immune mice increased and was always greate r than that in LPS-untreated immune control mice. Th e profile of the PFC response in L-LPS-treated mice was similar to that seen in the E-LPS-treated rmce. Enhancement of non -specific resistance of mice to Salmo nella infection. The mice treated with 2.5 ug L-LPS were effectively protected again st S. typhimurium (Fig. 7) ; no deaths occurred during the observation period. However, S. typhimurium was dete cted from the spleen, liver and other organs. The mice treated with 2.5 ug E-LPS were also protected but the enha ncement of non-specific resistance was less than that seen in L-LPS. Nine of the 10 controls died within 66 h. Similarly, mice showed nonspecific resistance by treatment with other doses of L-LPS (0.25 ug, 25 ug, 25 ug and 1000 ug) and E-LPS (0.25 !J.g and 25 ug). Activation of mouse peritoneal ma crophages. A strong phagocytic activity was seen in macrophages from mice treated with L-LPS; after 120 min incubation, S. typhimurium could not gro w on the agar plate (Fig. 8). Phagocytic activity was also 7

100 i<=.-a=-c..... ::=..C-_.......

.....e ~

8

6

5

24 42 48 66 90 114 138 HOURS AFTER INFECTION

o

30

60

90

120

INCUBATION TIME (min)

Fig. 7. Survival of L-LPS treated mice after intraperitoneal infection of S. typhimurium. Symbols, e, 2.5 ug of L-LPS; 0 , 2.5 ug of E-LPS; . , control. Fig. 8. Enhancement of bactericidal action in macrophages from mice inoculated with LPS. Symbols, e, macrophages from mice inoculated with 2.5 ug of L-LPS; macrophages from mice inoculated with 2.5 flg of E-LPS ; . , macrophages from control mice.

°

Biological Activities of Leptospiral Lipopolysaccharide

61

increased in macrophages from mice treated with E-LPS although the activity was lower than that of macrophages from mice treated with L-LPS. Limulus test. Concentration of more than 1 ng/ml L-LPS produced a positive test just like E-LPS. Chemical composition of L-LPS. The extraction procedure yielded L-LPS accounting for approximately 2.0 % of the weight of the cells. The final product contained 54 % carbohydrate, 12.2 % lipid and 5 % protein. Hexose, 6-deoxyhexose, pentose and hexosamine were 16 %,12 %, 22 % and 5 % respectively. 2-keto-3-deoxyoctanate (KDO) and uronic acid were not detected. The neutral sugars observed were arabinose (1.000), rhamnose (0.899), xylose (1.012), galactose (0,610), mannose (0.421) and fucose (0.610). The amount of each sugar was expressed as a molar ratio to arabinose. These sugars were identified as trimethylsilyl derivatives of their methyl glucosides. Serological specificity of L-LPS. As shown in Table 1, anti L-LPS antiserum agglutinated serovars belonging to Icterohaemorrhagiae serogroup, while the antiserum did not agglutinate 6 serovars (autumnalis, australis, pomona, hebdomadis, canicola and pyrogenes) of different serogroups. The L-LPS was tested by immunodiffusion against antisera to various serovars of Leptospira. The L-LPS formed a precipitin band with each of antisera against serovars belonging to Icterohaemorrhagiae, Sarmin and Smithi serogroup. Therefore, L-LPS appears to be a serogroup-associated antigen. Table 1. Serological examinations Serogroup

Serovar

Icterohaemorrhagiae copenhageni icterohaemorrhagiae mankarso naam

a

b

Formation of precipitin line with L-LPS and antiserum against each serovar

4096 4096 4096 4096

+ + + + +

sarmm

2048 2048

NT"

birkini mwogolo bogvere ndambari

2048 4096 4096 4096

+ +

ndhambukje

Others"

Agglutinin titers against anti-L-LPS antiserum

- «

NT NT 2)

Not tested. Six serovars (autumnalis, australis, pomona, hebdomadis, canicola and pyrogenes) of different serogroups.

Discussion L-LPS described in the present study showed a number of biological properties often associated with E-LPS endotoxin. Comparison of these properties is shown in Table 2. Some of biological activities of L-LPS are remarkably lower than those of E-LPS.

62

E.I sogai, H. Isogai, Y. Kurebayashi, and N. Ito Table 2. Comparison of biological activity between L-LPS and E-LPS Biological activity

L-LPS

E-LPS

Toxic effects Lethality Pyrogenicity Weight loss Bone marrow reaction Damage of the liver at day 1 Damage of the liver at day 7

+ + + + + +

+ + + + +

+

+

+ +

+ +

+

+ + +

Immunological effects Adjuvant effect Enhancement of non-specific resistance to heterogenous bacteria Activation of macrophages Others Limulus test Presence of KDO Presence of carbohydrate

+

This finding is similar to those observed in LPS fro m other bact eria. For instance, the activities of Yersinia pestis LPS in the limulus test and pyrogenicity studies was abo ut 10 times less potent than E-LPS on a weight basis (3). Capnocytophaga sputigena LPS was also found to be on ly weakl y po tent in comparison with ente robacterial endotoxin (23). Previous attempt s have been unsuccessful to demonstr ate an endo toxin-like activity in leptos pira l LPS (1, 8, 9). Th is is the first repo rt of biological activities associated with L-LPS. The differences between prior findings and the pr esent ones may by att ributed to the following . Firstly, biological activities can genera lly vary considerably depending upo n such factors as bacterial strains, growth cond itions of bacteria and ext raction procedures, as well as the species and age of experimental animals used for biological assay. Secondly, toxic effect of L-LPS was weak . For instance, L-LPS was active in lethality for mice but abo ut 12 times less po tent than E-LPS. It was reported that hemorrhage and necrosis in the bone marrow of mice could be used as a bioassay for E. coli endotoxin activities (29, 30 ). We obse rved hemorrhagic necrosis in the bone marrow after int ravenous injection of L-LPS an d th at the bone marrow was more vulnerable than other organs in the hemor rhagic response. Apparently, this method could be a useful too l as bioassay for L-LPS. Th e pathogenesis of jaun dice in leptospirosis has been attributed to a variety of mechanisms but none of them has yet been confirmed. It was suggested tha t a toxic substance, pro duced by lepto spiras or released fro m lepto spiras by lysis, is responsible for hepatocellular damage and then jaundice (20). In this study, hepatocellular dam age was demonstrated in the mice treated with L-LPS. The toxic substance for liver cells may be L-LPS or materials produced in response to L-LPS. L-LPS ind uced non -specific acquired cellular resistance to S. typhimurium. The enhanced bactericidal activity is considered to be due to macrophage activatio n. Entero bacterial LPS has been known to activate macro phages whic h inhibit th e growth of the homologou s organisms as well as antigenically unr elated ones (4). L-LPS was

Biological Activities of Leptospiral Lipopolysaccharide

63

antigenically different from Salmonella antigens (data not shown) but strongly induced macrophage response which inhibited Salmonella infection. This fact is interesting because macrophages are important in the host immune system. Apparantly L-LPS is much less toxic than E-LPS but does appear to stimulate lymphocytes and macrophages. The latter activities could be due to the protein (5 %) present in the L-LPS. At least proteins associated with LPS from enteric organisms have been reported to be potent activators of immunocompetent cells (25). KDO was not detected either in LPS or leptospiral extracts (8, 19). Zeigler and Vaneseltine failed to detect KDO in serovar pomona (33). KDO-undetectable LPS has also been isolated from Bacteroides (12, 13) and Vibrio (11). These LPS showed week endotoxic acitvities in contrast to classical LPS such as E-LPS. Further studies of their activities combined with the specific structures will be necessary. Serological tests indicate that the L-LPS is serogroup-associated but not serovarspecific. Serovar-specific lipopolysaccharide antigens (TM antigens) have been extracted from different serovars of Leptospira (16,21,27). However, it was doubtful to consider that the TM antigens were serovar-specific. Anti-TM antiserum reacted with different serovars and monoclonal antibodies against TM antigen also reacted with different serovars in their laboratory (18). Further studies will be necessary to test whether the endotoxin protein or other distinct structure components of leptospiras are responsible for the biological effects. Such structure may participate directly or indirectly in feature of leptospiral infection. References 1. Arean, V. M., G. Sarasin, and]. H. Green: The pathogenesis of leptospiras. Toxin production by Leptospira icterohaemorrhagiae. Amer. J. vet. Res. 25 (1964) 836-843 2. Britter, T. and H. M. Muir: A modified monic acid carbazole reaction. Analyt. Biochem. 4 (1962) 330-334 3. Butler, T. and G. Moller: Mitogenic response of mouse spleen cells and gelation of limulus lysate by lipopolysaccharide of Yersinia pestis and evidence for neutralization of the lipopolysaccharide by polymyxin B. Infect. Immun. 18 (1977) 400-404 4. Cluff, 1. E.: Effects of endotoxins on susceptibility to infections. J. infect. Dis. 122 (1970) 205-215 5. Dische, Z.: Spectrophotometric method for the determination of free pentose and pentose in nucleotides. J. Bio!. Chern. 181 (1949) 379-392 6. Dische, Z. and 1. B. Shettles: A specific color reaction of methylpentose and a spectrophotometric micromethod for their determination. J. Bio!. Chern. 175 (1948) 595-603 7. Dubois, M., K. A. Gilles, ]. K. Hamilton, P. A. Robers, and F. Smith: Colorimetric method for determination of sugar and related substances. Analyt. Chern. 28 (1956) 350-356 8. Paine, S., B. Adler, and A. Palit: Chemical and biological properties of a serotypespecific polysaccharide antigen in Leptospira. Aust. J. expo Biol, med. Sci. 52 (1974) 311-319 9. Pineo, D. R. and D. G. Low: Endotoxin properties of Leptospira canicola. Amer.]. vet. Res. 28 (1967) 1863-1872 10. Gardell, S.: Separation on Dowex 50 ion exchange resin of glucosamine and galactosamine and their quantitative determination. Acta Chern. Scand. 7 (1953) 207-215 11. Hisatsune, K., S. Kondo, T. Iguchi, M. Mach ida, S. Asou, M. Inaguma, and F. Yamamoto: Sugar composition of lipopolysaccharides of Family Vibrionaceae. Microbio!. Immuno!. 26 (1982) 649-664

64

E.Isogai, H.Isogai, Y. Kurebayashi, and N. Ito

12. Hor stad, T.: Biological activities of endotoxins from Bacteroides melaninogeni cus. Arch. Oral. BioI. 15 (1970) 343-348 13. Horstad , T.: The distribution of heptose and 2-keto-3-deoxyoctanate in Bacteroidacea e. J. gen. Microbiol. 85 (1974) 314-320 14. [ern e, N. K. and A. A. No rdin: Plaque formation in agar by single antibody producing cells. Science (Wash.) 140 (1963) 405 15. Karkhanis , Y. D., J. Y. Ze tter, J. J. Jackson, and D. S. Carlo: New and improved microassay to determine 2 keto-3-deoxyoctanate in lipopolysaccharide of Gram negative bacteria. Analyt, Biochem. 85 (1978) 595-601 16. Kasai, N. and R. Yanagawa : Studies on the antigenic determinant group of the typespecific antigen of Leptospira canicola. Zbl. Bakt. Hyg., I. Abt, Orig. A 228 (1974) 369-377 17 . Lowry, a. H., N. J. Rosebrough, A. L. Farr, and R. J. Randall: Protein measurement with the Folin phenol reagent. J. BioI. Chern. 193 (1951) 265-275 18. On o, E., M. Naiki, and R. Yanagawa: Production and characterization of monoclonal antibodies to lipopolysaccharide antigen of Leptospira interrogans serovar kremastos and canicola. Zbl. Bakt. Hyg., I. Abt, Orig. A 252 (1982) 414-424 19. Palit, A. and P. M. Harrison : Immunochemistry of extract s from Lept ospira interrogans serotype hardjo. J. gen. Microbiol. 100 (1977) 249-256 20 . Sapp, W. J., I. H. Shiddique, C. S. Williams, and T. Graham: Histopathologic evaluation of livers of pregnant hamsters infected with Leptospira canicola. Amer. J. vet. Res. 41 (1980) 1288-1292 21. Shinagawa, M. and R. Yanagau/a: Isolation and characterization of a leptospiral typespecific antigen. Infect. Immun. 5 (1972) 12-19 22 . Stahlheim, a. H. V.: Virulent and avirulent Leptospires: Biochemical activities and survival in blood. Amer. J. vet. Res. 32 (1971) 843- 849 23. Stevens, R. H., M. N. McArthu r, A. N owotny, and B. F. Hammond: Biological and chemical character ization of endotoxin from Capnocy tophaga sputigena. Infect. Immun.27 (1980) 246-254 24 . Stuart, A. E. and A. E. Davidson: Effect of simple lipids on antibody formation after injection of foreign red cells. J. Path. Bact. 87 (1964) 305-316 25. Sulzer, B. M. and G. W. Goodman: Endotoxin protein: a B-cell mitogen and polyclonal activator of C3H1HeJ lymphocytes. J. expo Med. 144 (1976) 821-827 26 . Trevelyan, W. E. and J. S. Harrison : Studies on yeast metabolism. I. Fraction and microdetermination of cell carbohydrates. Biochem. J. 50 (1952) 298-303 27 . Tsuji, M. , Y. Kawaok a, M. Nai ki, and R. Yanagawa: Isolation of antigenically active components from leptospiral serovar-specific lipopolysaccharide antigen by alkaline treatment. Microbiol. Immunol. 25 (1981) 949- 957 28. We stphal, a., a. Liideritz, and F. Bister: Uber die Extr aktion von Bakterien mit Phenol! Wasser. Z. Naturforsch. 76 (1952) 148-155 29. Yoshida , M., M. Hirata, Y. Hantano, and K. Inada: Hemorrh agic necrosis in mouse bone marrow induced by single injection of endotoxin, and application to the bioassay of endotoxin and to study on biological activitiy of K-antigen. Jpn. J. expo Med. 38 (1968) 335-346 30. Yoshida, M., M. Hirata, and K. Inada: Hemorrhage and necrosis in mouse bone marrow induced by endotox in - A method assaying quantiative changes of the cellularities. Jpn.J. expo Med. 43 (1973) 393-402 3 1. Young , L. S., P. Stevens, and]. Ingram : Functional role of antibody against " core" glycolipid of Enterobacte riaceae. J. Clin. Invest. 56 (1975) 850-861 32. WHO : Current problems in leptospirosis research. WId Hlth Org. Tech. Rep. Ser. No. 380 (1967) 8 33 . Ze igler, J. A. and W. P. Vaneseltine: Isolation and chemical characterization of outer envelope of Leptospira pomona. Canad. J. Microbiol. 21 (1975) 1102-1112

Dr. Emiko lsogai, Dept. of Preventive Dentistry, Higashi Nippon Gakuen University, School of Dentistry, Ishikari Tobersu 061-02, Japan