- Email: [email protected]
Isolation and purification by high performance liquid chromatography of a tetanus toxin fragment (Fragment [A-B]) derived from mildly papain-treated toxin
Toxicon VoL 27, No. 9, pp. 1055-1057, 1989 . Printed in Great Britain.
0041-0101/89 $3.00+ .00 ® 1989 Pergamon Prem plc
ISOLATION AND PURIFICATION BY HIGH PERFORMANCE LIQUID CHROMATOGRAPHY OF A TETANUS TOXIN FRAGMENT (FRAGMENT [A-B]) DERIVED FROM MILDLY PAPAIN-TREATED TOXIN KumImto OzuTsuau, LEI DIAN-LIANG, NAKABA SUGimoTo and MOR1HIRO MATSUDA* Department of Tuberculosis Research I, Research Institute for Microbial Diseases, Osaka University, 3-1 Yamadaoka, Suita, Osaka 565, Japan (Acceptedforpublication 8 March 1989)
K, OzuTsuhu, LEI DIAN-LIANG, N. SUGimoTo and M. MATSUDA. Isolation and purification by high performance liquid chromatography of a tetanus toxin fragment (Fragment [A-BD derived from mildly papain-treated toxin. Toxicon 27, 1055-1057, 1989 .-Fragment [A-B] of tetanus toxin was highly purified by combination of gel permeation chromatography, adsorption chromatography in an HPLC and immunoadsorption chromatography using anti-Fragment [C] as a ligand. The purified Fragment [A-B] (200 pg) elicited a peculiar toxicity, `hypoactivity' or `weakness', and killed the mice in ca . 73 hr and 88 hr when it was injected i.v. and i.m., respectively . However, contamination by the whole toxin was not detectable, in the purified fragment preparation, when up to 600,ug was tested by the mouse toxicity assay.
FRAGMENT [A-B] (an N-terminal fragment) of tetanus toxin which is obtained by digesting the toxin with papain (EC 3. 4. 4. 10) has been reported to elicit flaccid paralysis (HELTING et al ., 1978 ; MATSUDA et al., 1982a; GAWADE et al., 1985). However, SIMI'SON and HOCH (1985) showed that Fragment [A-B] did not block neuromuscular transmission and ascribed the toxicity apparently associated with the fragment to trace contamination with the whole tetanus toxin molecule . Thus, it is controversial whether Fragment [A-B] per se has a toxic effect . The inconsistency on the properties of Fragment [A-B] is mainly due to difficulty in removing extremely small amounts of undigested whole toxin from a preparation of Fragment [A-B]. In this study, we established a reproducible method for isolation and purification of Fragment [A-B] free from residual undigested whole toxin and then we showed that Fragment [A-B] carries the active domain(s) of tetanus toxin responsible for eliciting neurotoxicity. Nicked tetanus toxin, which was purified from culture filtrates according to the method previously described (OzuTSumi et al ., 1985a), was used for preparing Fragment [A-B]. Fragment [A-B] was prepared and partially purified by gel permeation chromatography in 'Author to whom correspondence should be addressed. 1055
1056
Short Communications
a high performance liquid chromatographic system as described previously (OzuTsum1 et al., 1985b). Contaminating whole toxin was 3.8-5 .4% of this partially purified fragment on a protein basis, as calculated from the toxicity (104 MLD per lug of toxin protein) of the whole toxin, which was estimated by mouse toxicity assay as described previously (MATSUDA et al., 1982b) . Subsequently, the fragment was applied to a KB column (Koken, Tokyo) equilibrated with 20 mM NaK phosphate buffer, pH 7.5, in an FPLC system (Pharmacia, Uppsala). Elution was performed by stepwise increase in the concentration of NaK phosphate buffer, pH 7.5, at a flow rate of 1 ml/min, as indicated in Fig. 1. The toxin contaminant in Fragment [A-B] was eluted as the first small peak (1 .5-4 .8% of the total amount of protein applied) (Fig. 1) using 40mM NaK phosphate buffer . The second peak (Fig. 1) that was eluted at 100 mM NaK phosphate buffer was identified antigenically by immunodiffusion as that of Fragment [A-B] (data not shown) . This step was effective for rapid removal of contaminating toxin and for concentrating the preparation of the fragment, but mouse toxicity assay showed that the purified preparation after this step was still contaminated with 0.5-2% of whole toxin molecules on a protein basis. The Fragment [A-B] was further purified by passing through an immunoadsorbent column in which rabbit antibody against purified Fragment [C], complementary fragment to Fragment [A-B] in the tetanus toxin molecule, was used as a ligand . Final recovery of Fragment [A-B] by the method described here was 27-28% of that in the whole toxin before papain digestion. On SDS-gel electrophoresis, the purified Fragment [A-B] migrated as a single protein band in a'position corresponding to a mol.wt of 96,000 f 5000. The reduced sample of the fragment migrated as two protein bands close in positions corresponding to mol.wts of 47,0000 t 2000 and 49,000 f 2000. On isoelectric focusing, Fragment [A-BI formed a single major band (pI5.52) with several minor bands (pI5.35-5.86), indicating the existence of microheterogeneity in Fragment [A-B] molecules. On immunoblotting using anti-Fragment [C] antibody, Fragment [C] and the whole toxin molecule were not detectable in the final purified Fragment [A-B] (data not shown) . The toxicities of tetanus toxin and Fragment [A-B] are summarized in Table 1 . The mice injected with 200 hg of Fragment [A-B] showed no symptoms for 48 hr, but then, they gradually became ill and died at 73 hr, as described by HELT7NG et al. (1978) . Even upon increasing the dose of injection to 600 hg of Fragment [A-B] per mouse, the mice did 200 O
N Ô a
0 .10 100
0 .05 0.00
10
20 30 Elution tlnn (min)
40
50
a 0
40 b[ 20-4 0
FIG. 1 . ELUTION PRoFuz of PARTIALLY PURIFIED FRAGMENT [A-B] ON A HYDROXYAPATrm KB coLumN (0.6 x 10 cm) coNNEcTED wrrH A PRECOLUMN (0.6 x 3 cm) . Buffer, NaK phosphate buffer, pH 7.5 . The concentration of the phosphate buffer for stepwise elution is indicated by the dotted line. Flow rate, 1 ml/min . Sample : 0 .13 mg of protein.
1057
Short Communications TARIE 1 . Toiac9TY of TETANUS ToxiN AND FRAGm3ENT [A-B] Time to death mean ±S .D. (hr) Symptoms
Route of administration
n
250 pg
i.m . i.v.
4 4
72±2 .8 70±1 .9
Delayed, with spastic paralysis
200 ug
i.v .
4
0.85±0.12
Acute, with flaccid paralysis
Fragment 200 leg [A-B] 200 ug
i.m . i.v .
4 8
88±4.5 73±9.3
Sample
Tetanus toxin
Dose
Delayed, with 'hypoactivity' or weakness
not show any symptoms for 48 hr. The symptom they showed after 48 hr was peculiar `hypoactivity' or `weakness' rather than typical flaccid paralysis. They never showed spastic paralysis. Recently TAKANo et al. (1988), in collaboration with us, showed electrophysiologically that intraspinal injection of the purified Fragment [A-B] obtained by the present method blocked both the inhibitory and excitatory synapses almost simultaneously, unlike the whole toxin which blocked the inhibitory synapses first and then the excitatory synapses. The peculiar symptoms of the mice may be explained by simultaneous blocking of both types of synapses. In case of whole tetanus toxin intoxication, two types of paralyses are known: (1) acute, flaccid (botulinum-like) paralysis at high doses, and (2) delayed, spastic paralysis at low doses (MATSUDA et al., 1982b) . The toxicity elicited by Fragment [A-B] was delayed but not spastic. Therefore the toxicity was not due to the trace amount of tetanus toxin contaminating the preparation but due to the Fragment [A-B] per se. REFERENCES GAwADE, S., BoN, C. and Btzzim, B. (1985) The use of antibody F(ab) fragments specifically directed to two different complementary parts of the tetanus toxin molecule for studying mode of action of the toxin. Brain
Res . 334, 139-146.
HELTING, T. B., RoNNEBERGEt, H. J., Vot.1ERTHUN, R. and NEUHAUER, V. (1978) Toxicity of papain-digested tetanus toxin, pathological effect of fragment B in the absence of spastic paralysis . J. biol. Chem. 253, 125-129 . MATSUDA, M., SuGamro, N. and Oztrtmm, K. (1982x) Acute botulinum like intoxication by tetanus in mice and the localization of the acute toxicity in the N-terminal papain fragment of the toxin. In: Proceedings of the 6th International Conference on Tetams, Lyon, 1981, pp. 21-32 . Lyon : Foundation Mérieux. MATSUDA, M., Summoro, N., OzuTSwa, K. and HuLm, T. (1982b) Acute botulinum-like intoxication by tetanus neurotoxin in mice . Blochem . biophys. Res. Commun. 104, 799-$05 . Ozursm, K., SUGImoTo, N. and MATSUDA, M. (1985x) Rapid, simplified method for production and purification of tetanus toxin. Appl. envir . Microbiol. 49, 939-941 . Oztnsubu, K., Summoro, N. and MATSUDA, M. (1985b) Rapid and simplified method of tetanus toxin and toxin fragments determination by HPLC. In : Seventh International Conference on Tetanus, pp . 49-56 (Ntsne6, G., MASrRoEm, P. and Pr zuRRA, M., Eds). Rome : Gangemi. Stmrsom L. L. and Hoctt, D. H. (1985) Neurophanmacological characterization of fragment B from tetanus toxin. J. Pharmac . exp . Ther. 232, 223-227 . TAKANO, K., MATSUDA, M., OzuTsum, K., KutctNER, F., GREmwt.T, A. and SUmmaro, N. (1988) Biological activity of a tetanus toxin fragment : action on the excitation and inhibition of the motor cell of the spinal cord . In: Proceedings of the 8th International Conference on Tetams, Leningrad, 1987, pp . 139-145 (Nwnc6, G., Btzztm, B., BYr(atENKO, B. and TRIAD, R., Eds) . Rome: Pythagora .
0041-0101/89 $3.00+ .00 ® 1989 Pergamon Prem plc
ISOLATION AND PURIFICATION BY HIGH PERFORMANCE LIQUID CHROMATOGRAPHY OF A TETANUS TOXIN FRAGMENT (FRAGMENT [A-B]) DERIVED FROM MILDLY PAPAIN-TREATED TOXIN KumImto OzuTsuau, LEI DIAN-LIANG, NAKABA SUGimoTo and MOR1HIRO MATSUDA* Department of Tuberculosis Research I, Research Institute for Microbial Diseases, Osaka University, 3-1 Yamadaoka, Suita, Osaka 565, Japan (Acceptedforpublication 8 March 1989)
K, OzuTsuhu, LEI DIAN-LIANG, N. SUGimoTo and M. MATSUDA. Isolation and purification by high performance liquid chromatography of a tetanus toxin fragment (Fragment [A-BD derived from mildly papain-treated toxin. Toxicon 27, 1055-1057, 1989 .-Fragment [A-B] of tetanus toxin was highly purified by combination of gel permeation chromatography, adsorption chromatography in an HPLC and immunoadsorption chromatography using anti-Fragment [C] as a ligand. The purified Fragment [A-B] (200 pg) elicited a peculiar toxicity, `hypoactivity' or `weakness', and killed the mice in ca . 73 hr and 88 hr when it was injected i.v. and i.m., respectively . However, contamination by the whole toxin was not detectable, in the purified fragment preparation, when up to 600,ug was tested by the mouse toxicity assay.
FRAGMENT [A-B] (an N-terminal fragment) of tetanus toxin which is obtained by digesting the toxin with papain (EC 3. 4. 4. 10) has been reported to elicit flaccid paralysis (HELTING et al ., 1978 ; MATSUDA et al., 1982a; GAWADE et al., 1985). However, SIMI'SON and HOCH (1985) showed that Fragment [A-B] did not block neuromuscular transmission and ascribed the toxicity apparently associated with the fragment to trace contamination with the whole tetanus toxin molecule . Thus, it is controversial whether Fragment [A-B] per se has a toxic effect . The inconsistency on the properties of Fragment [A-B] is mainly due to difficulty in removing extremely small amounts of undigested whole toxin from a preparation of Fragment [A-B]. In this study, we established a reproducible method for isolation and purification of Fragment [A-B] free from residual undigested whole toxin and then we showed that Fragment [A-B] carries the active domain(s) of tetanus toxin responsible for eliciting neurotoxicity. Nicked tetanus toxin, which was purified from culture filtrates according to the method previously described (OzuTSumi et al ., 1985a), was used for preparing Fragment [A-B]. Fragment [A-B] was prepared and partially purified by gel permeation chromatography in 'Author to whom correspondence should be addressed. 1055
1056
Short Communications
a high performance liquid chromatographic system as described previously (OzuTsum1 et al., 1985b). Contaminating whole toxin was 3.8-5 .4% of this partially purified fragment on a protein basis, as calculated from the toxicity (104 MLD per lug of toxin protein) of the whole toxin, which was estimated by mouse toxicity assay as described previously (MATSUDA et al., 1982b) . Subsequently, the fragment was applied to a KB column (Koken, Tokyo) equilibrated with 20 mM NaK phosphate buffer, pH 7.5, in an FPLC system (Pharmacia, Uppsala). Elution was performed by stepwise increase in the concentration of NaK phosphate buffer, pH 7.5, at a flow rate of 1 ml/min, as indicated in Fig. 1. The toxin contaminant in Fragment [A-B] was eluted as the first small peak (1 .5-4 .8% of the total amount of protein applied) (Fig. 1) using 40mM NaK phosphate buffer . The second peak (Fig. 1) that was eluted at 100 mM NaK phosphate buffer was identified antigenically by immunodiffusion as that of Fragment [A-B] (data not shown) . This step was effective for rapid removal of contaminating toxin and for concentrating the preparation of the fragment, but mouse toxicity assay showed that the purified preparation after this step was still contaminated with 0.5-2% of whole toxin molecules on a protein basis. The Fragment [A-B] was further purified by passing through an immunoadsorbent column in which rabbit antibody against purified Fragment [C], complementary fragment to Fragment [A-B] in the tetanus toxin molecule, was used as a ligand . Final recovery of Fragment [A-B] by the method described here was 27-28% of that in the whole toxin before papain digestion. On SDS-gel electrophoresis, the purified Fragment [A-B] migrated as a single protein band in a'position corresponding to a mol.wt of 96,000 f 5000. The reduced sample of the fragment migrated as two protein bands close in positions corresponding to mol.wts of 47,0000 t 2000 and 49,000 f 2000. On isoelectric focusing, Fragment [A-BI formed a single major band (pI5.52) with several minor bands (pI5.35-5.86), indicating the existence of microheterogeneity in Fragment [A-B] molecules. On immunoblotting using anti-Fragment [C] antibody, Fragment [C] and the whole toxin molecule were not detectable in the final purified Fragment [A-B] (data not shown) . The toxicities of tetanus toxin and Fragment [A-B] are summarized in Table 1 . The mice injected with 200 hg of Fragment [A-B] showed no symptoms for 48 hr, but then, they gradually became ill and died at 73 hr, as described by HELT7NG et al. (1978) . Even upon increasing the dose of injection to 600 hg of Fragment [A-B] per mouse, the mice did 200 O
N Ô a
0 .10 100
0 .05 0.00
10
20 30 Elution tlnn (min)
40
50
a 0
40 b[ 20-4 0
FIG. 1 . ELUTION PRoFuz of PARTIALLY PURIFIED FRAGMENT [A-B] ON A HYDROXYAPATrm KB coLumN (0.6 x 10 cm) coNNEcTED wrrH A PRECOLUMN (0.6 x 3 cm) . Buffer, NaK phosphate buffer, pH 7.5 . The concentration of the phosphate buffer for stepwise elution is indicated by the dotted line. Flow rate, 1 ml/min . Sample : 0 .13 mg of protein.
1057
Short Communications TARIE 1 . Toiac9TY of TETANUS ToxiN AND FRAGm3ENT [A-B] Time to death mean ±S .D. (hr) Symptoms
Route of administration
n
250 pg
i.m . i.v.
4 4
72±2 .8 70±1 .9
Delayed, with spastic paralysis
200 ug
i.v .
4
0.85±0.12
Acute, with flaccid paralysis
Fragment 200 leg [A-B] 200 ug
i.m . i.v .
4 8
88±4.5 73±9.3
Sample
Tetanus toxin
Dose
Delayed, with 'hypoactivity' or weakness
not show any symptoms for 48 hr. The symptom they showed after 48 hr was peculiar `hypoactivity' or `weakness' rather than typical flaccid paralysis. They never showed spastic paralysis. Recently TAKANo et al. (1988), in collaboration with us, showed electrophysiologically that intraspinal injection of the purified Fragment [A-B] obtained by the present method blocked both the inhibitory and excitatory synapses almost simultaneously, unlike the whole toxin which blocked the inhibitory synapses first and then the excitatory synapses. The peculiar symptoms of the mice may be explained by simultaneous blocking of both types of synapses. In case of whole tetanus toxin intoxication, two types of paralyses are known: (1) acute, flaccid (botulinum-like) paralysis at high doses, and (2) delayed, spastic paralysis at low doses (MATSUDA et al., 1982b) . The toxicity elicited by Fragment [A-B] was delayed but not spastic. Therefore the toxicity was not due to the trace amount of tetanus toxin contaminating the preparation but due to the Fragment [A-B] per se. REFERENCES GAwADE, S., BoN, C. and Btzzim, B. (1985) The use of antibody F(ab) fragments specifically directed to two different complementary parts of the tetanus toxin molecule for studying mode of action of the toxin. Brain
Res . 334, 139-146.
HELTING, T. B., RoNNEBERGEt, H. J., Vot.1ERTHUN, R. and NEUHAUER, V. (1978) Toxicity of papain-digested tetanus toxin, pathological effect of fragment B in the absence of spastic paralysis . J. biol. Chem. 253, 125-129 . MATSUDA, M., SuGamro, N. and Oztrtmm, K. (1982x) Acute botulinum like intoxication by tetanus in mice and the localization of the acute toxicity in the N-terminal papain fragment of the toxin. In: Proceedings of the 6th International Conference on Tetams, Lyon, 1981, pp. 21-32 . Lyon : Foundation Mérieux. MATSUDA, M., Summoro, N., OzuTSwa, K. and HuLm, T. (1982b) Acute botulinum-like intoxication by tetanus neurotoxin in mice . Blochem . biophys. Res. Commun. 104, 799-$05 . Ozursm, K., SUGImoTo, N. and MATSUDA, M. (1985x) Rapid, simplified method for production and purification of tetanus toxin. Appl. envir . Microbiol. 49, 939-941 . Oztnsubu, K., Summoro, N. and MATSUDA, M. (1985b) Rapid and simplified method of tetanus toxin and toxin fragments determination by HPLC. In : Seventh International Conference on Tetanus, pp . 49-56 (Ntsne6, G., MASrRoEm, P. and Pr zuRRA, M., Eds). Rome : Gangemi. Stmrsom L. L. and Hoctt, D. H. (1985) Neurophanmacological characterization of fragment B from tetanus toxin. J. Pharmac . exp . Ther. 232, 223-227 . TAKANO, K., MATSUDA, M., OzuTsum, K., KutctNER, F., GREmwt.T, A. and SUmmaro, N. (1988) Biological activity of a tetanus toxin fragment : action on the excitation and inhibition of the motor cell of the spinal cord . In: Proceedings of the 8th International Conference on Tetams, Leningrad, 1987, pp . 139-145 (Nwnc6, G., Btzztm, B., BYr(atENKO, B. and TRIAD, R., Eds) . Rome: Pythagora .