Vol. 24, No. 3

INFECTION AND IMMUNITY, June 1979, p. 958-961 0019-9567/79/06-0958/04$2.00/0

Toxin-Neutralizing Effect of Antibody Against SubtilisinDigested Tetanus Toxin HIROKO SATOt AKIHARU ITO, YOSHIO YAMAKAWA, AND RYOSUKE MURATA

Second Department of Bacteriology, National Institute of Health, Kamiosaki, Shinagagawa-ku, Tokyo 141, Japan Received for publication 29 March 1979

A form of systemic tetanus with atypical symptoms was observed in mice injected in the left thigh with a mixture of tetanus toxin and antibody produced in guinea pigs against a fragment of toxin obtained from a subtilisin digest of the crystallized toxin. The mice did not show typical symptoms of the local tetanus such as convulsions or spastic paralysis of the injected limb.

Several studies have shown that extracellular tetanus toxin consists of two polypeptide chains, heavy and light, with molecular weights of about 100,000 and 50,000, respectively. These fragments result from proteolytic nicking of intracellular toxin composed of a single polypeptide chain with a molecular weight of about 150,000 (1, 2, 6). On the basis of immunochemical studies with antibodies to fragments obtained by treating tetanus toxin with proteases or denaturants, the structural relationship among the fragments was clarified, and a schematic model of antigenic structure of the toxin was proposed (1, 4, 6, 7, 9). However, there are only a few reports concerning the relationship of toxin fragments to the mechanism of action of the toxin. Helting and Zwisler (4), using papain digestion of the toxin, isolated fragment C (molecular weight, 50,000) and fragment B (molecular weight, 100,000) and subsequently reported (5) that a binding site for ganglioside was located on the heavy chain of the toxin, possibly in or near the region comprised by fragment C. Helting et al. (3) found that mice injected with a large amount of fragment B exhibited a peculiar retraction of the abdominal muscles, but never gave the typical features of tetanus. We now report that tetanus without local symptoms was observed in mice injected into the left thigh with a mixture of tetanus toxin and antibody against a 50,000-molecular-weight fragment (named fragment S) obtained after subtilisin digestion of the tetanus toxin. This fragment S may correspond to fragment C as described above (3, 4). Tetanus toxin was purified from a culture filtrate of Clostridium tetani (Harvard strain), grown on a modified Mueller and Miller met Present address: Division of Bacterial Products, Bureau of Biologics, Food and Drug Administration, Bethesda, MD 20205.

dium. The culture filtrate was obtained from Takeda Chemical Ind., Ltd. (Hikari, Japan). The culture filtrate concentrated by ultrafiltration through Hollow Fiber (Amicon HIDP 10) was fractionated with ammonium sulfate between 20 and 40% saturation. The fractionated toxin was applied to a column of DE 52 (Whatman) equilibrated with 0.05 M tris(hydroxymethyl)aminomethane-hydrochloride (Tris-hydrochloride) buffer, pH 7.5. The column was washed with several column volumes of the buffer, and then the toxin was eluted with the buffer containing 0.03 M NaCl. Both activity and protein were found in two adjacent peaks. The two active fractions were pooled separately, and each of these purified fractions showed single bands with identical electrophoretic mobilities in disc electrophoresis. They also had almost the same specific activity in flocculating units per absorbance at 280 nm (A280) and 50% lethal dose per A280. Crystallization of the purified toxin was done with each of the above active fractions. The toxin was precipitated with 40% saturation of ammonium sulfate. The precipitate was dissolved in a minimal amount of 0.05 M phosphate buffer (pH 7.5). The concentrated toxin was dialyzed against the same buffer and then centrifuged to remove some aggregates. The clear supernatant fluid was processed for crystallization. Finely powdered ammonium sulfate was carefully added to the toxin solution (A280 = 23) until a faint turbidity appeared (about 25% saturation of ammonium sulfate). A few drops of saturated ammonium sulfate was added slowly with stirring, and the mixture was placed in a refrigerator. One or two drops of saturated ammonium sulfate solution was carefully added every morning and evening. On storage of this preparation in a refrigerator, needle- or spindleshaped white crystals of tetanus toxin were ob-

958

VOL. 24, 1979

tained within 20 days. For recrystallization, the crystals were dissolved in the minimal amount of 0.05 M phosphate buffer (pH 7.5). The powdered ammonium sulfate and saturated ammonium sulfate solution were carefully added as described above, and a few seed crystals of the toxin were added. The mixture was kept in the refrigerator. The crystalline tetanus toxin of about 20 ,um in length was obtained within 3 days (Fig. 1). The crystalline toxin was very stable and could be stored for more than 4 years in the refrigerator without change of molecular size (molecular weight, 150,000) and specific activity (307 flocculating units per Am0, 4.2 X 106 50% lethal dose per Am0). The crystalline toxins prepared from the two active fractions from the DE 52 column showed identical properties. The molecular size of the crystalline toxin was different from that reported by Pillemer and Moore (10). The fragment S was prepared by treating the crystallized tetanus toxin with subtilisin BPN' (subtilisin to the toxin, 1:20, at 370C for 4 h) in 0.05 M Tris-hydrochloride buffer (pH 8.0) and purified from the digest by successive gel filtration (0.05 M Tris-hydrochloride buffer containing 0.15 M NaCl, pH 7.5) on Sephadex G-150 and Sephadex G-100 superfine. The purified fragment S did not show any lethality against mice at an Am0 of 0.01, and its molecular weight was calculated to be 50,000 by gel filtration. As shown in Fig. 2, fragment S revealed three bands on polyacrylamide gel electrophoresis but a single precipitin line with incomplete or complete fusion with tetanus toxin by double diffusion test against guinea pig antitoxin (AT) or antifragment S (AF), respectively. To obtain antibodies against tetanus toxin and the fragment S,

~

~

~

~

I

4tSoAe.n

8' Of A;

v

_

E

to

>I

.,

X

-T

. \ 1

XJ

s -I

t

t

I

-.

t %

o

a

1%,

i~~~i

^

*'

t

*_ s

., S

I,, J

^

'.

,

/

w

I

_

_

0

e..

Ad.

I

I

I

e A, I.

t

@

a,'

>

-w4

959

guinea pigs were immunized with Formalintreated antigen with aluminum hydroxide gel as adjuvant. Guinea pigs were injected subcutaneously with the tetanus toxoid or the Formalintreated fragment S at A280 of 0.03 and 0.07, respectively. At 6 weeks after the first injection, each animal was given a second injection with half the amount of the first injection and bled 1 week after the second injection. The immunoglobulin fraction of the sera was collected with ammonium sulfate at 33% saturation. Determination of toxin-neutralizing activity of the antibodies was carried out at L+/10 and L+/1,000 levels in mice. Standard tetanus antitoxin calibrated against the international standard was used to calculate international units of the antitoxin. Table 1 summarizes the neutralizing effect of AT or AF on tetanus toxin in mice. The mixtures of diluted AT or AF and the purified toxin of the test doses were incubated for 1 h at room temperature and then injected into the left thigh of five mice. The international units of AT at L+/10 and L+/1,000 levels were calculated to be 1,180 and 584 IU/ml, respectively. The units of AF were calculated in a manner similar to those of AT, although the value may be less accurate because of a difference of avidity of AF; the titers at L+/10 and L+/1,000 levels were 313 and 51 IU/ml, respectively. Either antibody was able to neutralize the toxin in mice, but when doses of the antibody were insufficient to neutralize the test dose of the toxin completely, symptoms in mice injected with AF were different from those injected with AT. In the latter case, typical symptoms of local tetanus were observed, and the severity of symptoms seemed to depend on a dose of the free toxin in the AT-

_.

41 -P ~

NOTES

S

v

R 0

N

A

-

FIG. 1. Crystalline tetanus toxin (x600). Light and phase-contrast (inset) microphotographs.

960

NOTES

INFECT. IMMUN.

FIG. 2. Disc electrophoresis and double immunodiffusion test of crystallized tetanus toxin and fragment S. Symbols: T, tetanus toxin; F, fragment S; AT, guinea pig antitoxin; AF, guinea pig anti-fragment S.

Antibodies

Dose'

of the toxin

TABLE 1. Toxin neutralizing effect ofAT and AF Injected Severityb of symptoms at time dose after injection (day):

of the antibody

1

2

3

4

5

6

7

8

Symptom 9

10

None 1/7,590 Local 1 3C Deadd 1/11,390 Local 1/17,090 Dead None L+/1000 1/320,000 1 Local 1 1 2 2 2 2 2 2 1 1/640,000 Local 1/1,280,000 2 3 Dead None AF L+/10 1/900 ± ± ± ± ± Systemic 1/1,350 + + + + + + ++ ++ Systemic 1/2,030 c ++ Deadd Systemic 1/3,030 Local Dead 1/4,560 None L+/1000 1/10,000 ± ± ± ± ± Systemic 1/20,000 + + + ++ ++ +++ +++ +++ Dead Systemic 1/40,000 ++ Dead Systemic 1/80,000 1 3 Dead Local 1/160,000 aDose of the toxin at L+/10 and L+/1,000 levels contained 4,800 and 24 50% lethal dose mouse, respectively. b Severity of local tetanus is expressed in numerical values assigned by Mellanby et al. (8). The severity in the systemic tetanus was determined by rigidity, form, and condition of the mouse and is expressed in the signs from ± to +++. Five mice were used per dose, and each showed the same type and severity of symptoms. 'Photographs of this mouse are shown in Fig. 3. d Blood was taken out of these mice (see text). AT

L+/10

toxin mixture. On the other hand, when an insufficient amount of AF was used, mice developed two distinct types of symptoms. With low amounts of AF (1/4,560 group at L+/10 level, and 1/160,000 group at L+/1,000 level), mice died with the local tetanus; however, with larger amounts of AF (1/1,350, 1/2,030, and 1/3,030 groups at L+/10 level, and 1/20,000, 1/40,000 and 1/80,000 groups at L+/1,000 level), a systemic tetanus with symptoms which were very similar to the general tetanus produced by intravenous or intraperitoneal injection of tetanus toxin was observed.

In this systemic tetanus, spastic paralysis of the injected limb was never observed, and systemic rigidity, especially stiffness of abdominal muscle, weight loss, and delay of onset of the first symptoms were characteristics. Purified tetanus toxin was used as the test toxin in this experiment, but the same result was obtained when crude tetanus toxin was used. Fig. 3 shows typical symptoms of the local and the systemic tetanus produced by injection of the toxin with AT and AF, respectively. Blood was taken out of heart of these mice just after death as shown in Table 1. The toxic activity was detected in

NOTES

VOL. 24, 1979

961

FIG. 3. Photographs of mice injected with tetanus toxin plus AT (A and B) or AF (C and D). (A and B), Mice with typical local symptoms caused by non-neutralized excess toxin. (C and D), Mice do not show any local symptoms of tetanus. The amounts of antibody and toxin injected in these mice are given in Table 1 (footnote c).

the blood from the mice developing systemic tetanus but not in the others. It is quite possible that fragment S may correspond to fragment C described by Helting et al (3). Thus, the resemblance of the symptoms caused by the mixture of toxin and AF to that caused by fragment B (3) could be due to AF neutralizing that portion of the toxin corresponding to fragment C, but not neutralizing that portion corresponding to fragment B. Arguing against this, perhaps, is the fact that a large amount of AF was able to completely neutralize the toxin. Another factor to consider is the avidity of AF. At the injection site in the thigh, AF may neutralize the receptor binding site of the toxin by binding to fragment C (4). This would prevent local symptoms. But if the toxin-AF complex entered the blood, this complex might dissociate to allow free toxin to produce systemic tetanus. In any case, in considering the use of partial antigens as immunizing agents for tetanus, the understanding of the mechanism of development of systemic tetanus may be significant. UTERATURE CITED 1. Bizzini, B. 1977. Tetanus toxin structure as a basis for elucidating its immunological and neuropharmacologi-

2.

3. 4.

5. 6.

7. 8.

9.

10.

cal activities, p. 175-218. In P. Cuatrecasas (ed.), The specificity and action of animal, bacterial and plant toxins, series B, vol. 1. Chapman and Hall, London. Craven, C. J., and D. J. Dawson. 1973. The chain composition of tetanus toxin. Biochim. Biophys. Acta 317:277-285. Helting, T. B., H.-J. Ronneberger, R. Vollerthun, and V. Neubauer. 1978. Toxicity of papain-digested tetanus toxin. J. Biol. Chem. 253:125-129. Helting, T. B., and 0. Zwisler. 1977. Structure of tetanus toxin. I. Breakdown of the toxin molecule and discrimination between polypeptide fragments. J. Biol. Chem. 252:187-193. Helting, T. B., and 0. Zwisler. 1977. Structure of tetanus toxin. II. Toxin binding to ganglioside. J. Biol. Chem. 252:194-198. Matsuda, M., and M. Yoneda. 1975. Isolation and purification of two antigenically active, "complementary" polypeptide fragments of tetanus neurotoxin. Infect. Immun. 12:1147-1153. Matsuda, M., and M. Yoneda. 1977. Antigenic structure of tetanus neurotoxin. Biochem. Biophys. Res. Commun. 77:268-274. Mellanby, J., H. Mellanby, D. Pope, and W. E. van Heyningen. 1968. Ganglioside as a prophylactic agent in experimental tetanus in mice. J. Gen. Microbiol. 54: 161-168. Nagel, J., and H. Cohen. 1973. Studies on tetanus antitoxins. II. Demonstration of at least four antitoxins of different specificity in antitoxic sera. J. Immunol. 110: 1388-1395. Pillemer, L, and D. H. Moore. 1948. The spontaneous conversion of crystalline tetanal toxin to a flocculating atomic dimer. J. Biol. Chem. 173:427-428.