- Email: [email protected]
The in vitro inhibition of GABA release by tetanus toxin
NrurophurmocoiogyVol. 21, pp. 851 to 855, 1982
~28-3908/82~090851-05~3.~~0 Copyright 0 1982 Pergamon Press
Printed in Great Britain. All rights reserved
Ltd
THE IN VW-R0 INHIBITION OF GABA RELEASE BY TETANUS TOXIN G. L.
COLLINGRIDGE*
and J.
DAVIES
Department of Pharmacology, The School of Pharmacy, University of London, 29/39 Brunswick Square, London WCtN IAX, U.K. (Accepted 23 February 1982) in vitro effects of tetanus toxin on the potassium(K+)-evoked release of radiolabelled ;a-aminobutyric acid (GABA) from slices prepared from substantia nigra of the rat have been investigated. The addition of tetanus toxin (75(rlS,OOO mouse LD,,ml-‘) to the slices produced a dose- and time-dependent reduction in the K+-evoked release of C3H]GABA. These effects of the toxin were diminished at low tem~rature and in the absence of calcium ions in the bathing medium. It is suggested that this relatively simple in &ro system may allow more quantitative studies to be made concerning the action of the toxin on inhibitory amino acid release from the mammalian central nervous system.
Summary-The
Thousands of people are affected annually by tetanus and there is a high mortality rate since the mechanism of action of tetanus toxin is unknown and hence no specific treatment exists. The major locus of action of the toxin appears to be on the central nervous system (Brooks, Curtis and Eccles, 1957; Sherrington, 1906) where it has been shown to interfere with amino acidmediated synaptic inhibition by reducing the release of the transmitters, ;J-aminobutyric acid (GABA) and glycine, from nerve terminals (Collingridge, Collins, Davies, James, Neal and Tongroach, 1980; Curtis and de Groat, 1968; Davies and Tongroach, 1979; Guschin, Kozhechkin and Sverdlov, 1969: Osborne. Bradford and Jones, 1973). In these studies effects were achieved using either systemic or direct injection of the toxin into the whole animal. More information regarding the specificity and mechanism of action of the toxin in the CNS would probably be obtained if the toxin could be shown to be active following in vitro administration (cf experiments with botuIinum toxin, e.g. Simpson, 1980). Recently it has been reported in a preliminary form that the release of GABA from brain slices can be reduced by in citro administration of tetanus toxin (Colhngridge and Davies, 1980a; Collingridge, Thompson, Davies and Mellanby, 1981). Here the in vitro effects of tetanus toxin on C3H]GABA release from slices of rat substantia nigra are described in more detail.
were incubated in 2 ml oxygenated Krebs-Ringer bicarbonate solution (pH 7.4) at 37°C for an incubation period of 220min. Tetanus toxin (750-15,000 mouse LD5eml- ‘) was present for various periods (see Table legends) and C3H]GABA (8.7 x lo-‘M; 1 &i) was added for the last 30min of the incubation period. Slices were then transferred to small chambers and su~rfused with normal Krebs-Ringer bicarbonate solution at 37°C. After 24 min, consecutive 2 min fractions (2 ml) of the superfusate were collected, and at 30 min, release was evoked by increasing the K+ concentration of the Krebs-Ringer bicarbonate solution by 25 mM for 2 min. Aminooxyacetic acid (100pM) was present throughout to prevent metaboIism of C3H]GABA (Srinivasan, Neal and Mitchell, 1969) and the radioactivity released in each fraction was expressed as a percentage of tissue stores at that time. Evoked release was determined as the total release during and up to 10min after high K’ concentration minus the expected resting release at that time, which was calculated from the 8 min period preceding high K+ concentration, The release of C3H]GABA from slices incubated with toxin was compared, in parallel experiments, with the release from slices incubated under identical conditions in the absence of toxin or with tetanus toxin previously neutralized with tetanus antitoxin. RESULTS
METHODS The experimental methods have been described in detail elsewhere (Coilingridge er al., 1980). Briefly, 0.35 mm thick slices of substantia nigra from the rat * Present address for correspondence: Department of Physiology, The University of British Columbia, Vancouver, B.C., Canada V6T 1W5. Key words: tetanus toxin, substantia nigra, GABA, release.
Tetanus toxin had no effect on the spontaneous release of E3H]GABA from nigral slices, under any conditions employed. For example, following 210 min incubations in normal Krebs-Ringer bicarbonate solution, media containing active toxin (15,000 LD,e) or media containing neutralized toxin, the respective resting releases of 0.080 j, 0.004 (n = 25), 0.079 &- 0.009 (4) and 0.073 k 0.004 (4) “/, tissue stores min-’ were not significantIy different. Under appro-
852
G. L. COLLINGRIDGE and J. DAVIES
B
1
1 24
25 mM
1
I
26
-
I
k’
I
I
36
40
,
25 mM Time,
min
Fig. 1. Effect of tetanus toxin on the release of [jH]GABA from nigral slices. Each point is the mean + SEM of the release from four superfusion chambers containing slices (4-6 per chamber) exposed to tetanus toxin (dashed line) or control slices (continuous line). All eight superfusion chambers were run simultaneously. In (A) slices were incubated with 5000 mouse LD,, tetanus toxin or in the absence of toxin for 210 min before release experiments were commenced. In (B) slices were incubated for 210 min with 15,000 mouse LDso tetanus toxin or neutralized toxin (control).
priate conditions. however, active but not neutralized tetanus toxin, caused a significant reduction in the K+-evoked release of c3H]GABA from nigral slices (Fig. I). The effects of varying doses and exposure times of the slices to tetanus toxin on the release of C3H]GABA are summarised in Table I. Thus, with 210 min exTable 1. Effect of dose and exposure time with tetanus toxin on the K’-evoked release of C3H]GABA from nigral slices Interval of exposure to tetanus toxin (min) lc-220 lo-220 l&220 l&220 lo-40 l&70 16&220
Dose (mouse LDs, ml 750 2000 5000 15.000 5000 5000 5000
I)
Percentage reduction in K+-evoked release 12.2 25.1 51.7 66.0 7.7 45.9 8.8
k k + * k F +
7.4(6) NS 6.4(11)** 2.6(26)*** 3.7 (4) *** 8.5 (4) NS 9.2 (5)* 8.8(6)NS
Nigral slices were incubated for 220 min in Krebs solution. Tetanus toxin was present for the time shown in the left hand column. In experiments where toxin was added to the incubation media after 10 min and left in contact with the tissue for only 30 or 60min. slices were washed and incubations continued in normal Krebs solution for the remaining 180 and 150 min, respectively. Results are expressed as the mean + SEM percentage reduction in K+-evoked C3H]GABA release compared to the release in parallel control experiments. The figures in parenthesis refer to the number of pairs of superfusion chambers. Significance of difference (unpaired two-tailed l-tests) * P < 0.02: ** P < 0.01: *** P < 0.001.
posure to the toxin the effects were dose-dependent over the range examined (75&l 5,000 mouse LDsOmlm ‘). When toxin was present for the 10-70 min portion of the total 220min incubation, similar effects on release were seen, suggesting that the toxin had been irreversibly accumulated by the tissue within I hr. However. no significant effect was observed when the toxin was present for the 16&220 min period (Table 1) suggesting that an interval of time (greater than 30 min) must elapse after the toxin has been accumulated before appreciable effects develop. These results prompted a further experiment to investigate the time course of the effects of the toxin. Thus, slices were incubated in the presence or absence of tetanus toxin (5000 LDSo) for I hr before being washed and superfused for an additional 210min. Release was evoked 30 min after the start of superfusion in the normal manner, and at 60 min intervals thereafter: Figure 2 shows that 30 min after the slices had been removed from the toxin there was only a small reduction in K+-evoked GABA release (7.8”~ whereas 1 hr later there was a much greater effect (48.3”,, reduction). At this time the effects of the toxin appeared almost maximal since similar reductions were observed I and 2 hr later. Finally. preliminary studies were performed to investigate the effect of incubation temperature and of Ca2+ ions on the action of the toxin. When slices were incubated with tetanus toxin for ISOmin at 2.5 C and then at 37’ C for 30 min (while loading the slices with C3H]GABA) instead of at 37 ‘C throughout
853
Tetanus toxin and GABA release 0.B
r
Time,
min
Fig. 2. Time course of the effects of tetanus toxin on GABA release. Slices were incubated for 1 hr in the presence or absence of tetanus toxin (5000 mouse LDso) and then superfused for 222min. C3H]GABA release was evoked by 2 min K+ pulses (25 mm) at 30 min and hourly intervals thereafter. The 4 resulting release peaks are illustrated for control (continuous) and toxin-incubated slices (dashed line), but 40 min intervals between K’ pulses have been omitted. See text for further discussion of Figure. This experiment was performed twice and yielded similar results both times.
there was no change in the accumulation of C3H]GABA or its subsequent spontaneous release, However, there was a considerable reduction in the K+-evoked release (-70%; release performed at 37°C). When compared to parallel controls, the toxininduced reduction of K+-evoked release was smaller following incubation at low temperature (Table 2) suggesting that the action of the toxin may be temperature-dependent. Omitting Ca*+ ions from the media for the total 220 min incubation period did not significantly affect the uptake, spontaneous or evoked release of C3H]GABA (release being performed in the presence of Ca* ‘) but did reduce the effect of the toxin on the evoked release of C3H]GABA from nigral slices (Table 2).
The effects of tetanus toxin on K+-evoked C3H]GABA release cannot be attributed to a reduction in uptake since release is expressed as a percentage of tissue stores. Additional effects on uptake seem unlikely since the radioactivity remanning at the end of the experiment in control and toxin-incubated slices was never significantly different. (There was often slightly more radioactivity remaining in the toxinincubated slices which can be attributed to the smaller amount released from these slices). Furthermore, in separate uptake experiments no effects of the toxin were detected. For example, following incubation for 180 min with 2000 mouse LD,, tetanus toxin or in the absence of toxin the respective IO min tissue: medium ratios (dpm g- 1 tissue: dpm ml- I
Table 2. Effect of temperature and Ca ‘+ ions on the action of tetanus toxin
Temperature (“C) 37 2.5 37 37
Ca2+ ion cont. (mM) 2.52 2.52 2.52 0
K+-evoked release [‘HIGABA (“/, tissue stores) Control 5.77 + 1.69 + 4.13 5 4.90 f
0.79 0.13 0.33 0.54
Toxin
Percentage reduction in K “-evoked release of cJH]GABA
2.25 & 0.39 1.17 + 0.59 1.85 &-0.19 2.85 + 0.17
61.0 _t 6.7 (4) 30.6 k 3.5 (4)* 55.2 t 4.7 (6) 41.9 + 3.4(6)**
Nigral slices were incubated for 210 min with tetanus toxin (5000 mouse LD,, ml- ‘) and the mean k SEM release from these slices (fourth column) was compared with the release in parallel control experiments (third column). The fifth column presents the mean k SEM% reduction compared to the mean control value and the figures in parenthesis refer to the number of pairs of superfusion chambers. In the first series of experiments the release from slices (control and toxin) incubated for 180 min at 2.5”C and then 30min at 37°C was compared with slices incubated for 2lOmin at 37°C. In the second series, the effects of omitting Ca ‘+ ions from the incubation medium for 210 min was examined. Significance of difference in toxin-induced reduction in K+-evoked release. *P < 0.05;**p < 0.001.
G.
854
L.
COLLINGRIDGE and J. DAVIES
medium; n = 4) of 6.9 k 0.8 and 7.7 k 0.8 were not significantly different. DISCUSSION
In previous studies it was reported that tetanus toxin reduced Ca”-dependent K’-evoked release but not uptake, the small Ca”-independent K+-evoked release or the spontaneous release of 13H]GABA from nigral slices following in L’I’UO pretreatment with the toxin (Collingridge et al.. 1980). The present results confirm that tetanus toxin reduces the K+-evoked release of C3H]GABA from nigral slices in a Ca’ +-containing medium. It is likely, therefore, that the reduction in GABA-mediated synaptic inhibition in the nigra seen in ciw following intranigral injections of the toxin (Collingridge and Davies, 1980b, 198 1; Davies and Tongroach, 1979) is due to an effect on the synaptic release of GABA. The novel observation is that C3H]GABA release could be affected by in citro administration of the toxin under appropriate conditions, implying that the nigral slice contains all the necessary structures for tetanus toxin to act. Since in this slice preparation GABAergic terminals are largely or totally severed from their cell bodies (Gale and Iadarola, 1980) and since the K+-evoked release of C3H]GABA is probably from nerve terminals (Collingridge et N/., 1980). the toxin presumably affects release via an action directly on nerve terminals rather than indirectly via an effect on axons or cell bodies. The effects of tetanus toxin developed slowly. Thus, over 90min was required, after first exposure of the slices to the toxin, before a significant reduction in C3H]GABA release was detected (Table 1) and maximal effects took approx. 150min to develop (Fig. 2). This slow time course of the action of the toxin probably explains why no effect was observed on the electrically-evoked release of endogenous GABA or glycine only 30 min after in vitro administration of the toxin to synaptosomes from the rat brain (Osborne cut (II., 1973). Recently, Bigalke. Heller, Bizzini and Habermann (1981) have also reported an in vitro action of tetanus toxin on GABA release from the CNS. They observed similar reductions in [‘HIGABA release, to those reported here. following 2 hr incubations of rat brain particles with doses of toxin comparable to those used in the present study. Interestingly, in the present study it was only necessary for the toxin to be present for a relatively short period of time (greater than 30 min but less than 60 min) for full effects to develop, providing incubation of the slices was then continued in the absence of toxin. This observation suggests that the toxin is irreversibly accumulated by the slices within 1 hr and then an additional slow process is necessary before release is affected. The 3&60 min exposure time of the slices necessary for effects to develop subsequently was probably determined by the slow diffusion of the toxin molecules through the slices rather than the
actual binding process which is believed to be much more rapid. The effects of tetanus toxin appear not to be specific towards the release of inhibitory amino acid transmitters since the release of dopamine was reduced from slices prepared from toxin-poisoned striata (Collingridge et al.. 1980) and acetylcholine and noradrenaline release were decreased when brain particles were incubated with the toxin (Bigalke c’f I//.. 1981). Furthermore, acetylcholine release from primary nerve cell cultures (Bigalke, Dimpfel and Habermann. 1978) and from peripheral preparations (Habermann. Dreyer and Bigalke. 1980) was reduced following incubation with tetanus toxin. However, the release of GABA seems to be particularly sensitive to the action of tetanus toxin (Bigalke rt rrl.. 1981; Collingridge clr crl., 1980) and in the nigra is reduced by pretreatment with sublethal doses of the toxin (Collingridge and Davies. 1979). These and other data (Mellanby and Green. 1981) suggest that effects on GABA release are involved in tetanus. Therefore, this preparation should be useful in elucidating the pathogenic mode of action of the toxin and in the evaluation of drugs for specific antitetanus properties. Finally. this relatively simple system should enable quantitative studies to be made concerning the actions of the toxin on GABA release analogous to those performed with botuiinum toxin on acetylcholine release at the neuromuscular junction (Simpson, 1980). Ac,k~~o~~/rdyt,mr,nr,s~ We thank Dr M. Simmonds ing the manuscript Davey for technical
and Rachel assistance.
Williamson
and
for readRichard
KEFERENCES
Bigalke, H.. Dlmpfel. W. and Habermann. E. (1978). Suppression of ‘H-acetylcholine release from primary nerve cell cultures by tetanus and botulinurn-A toxin. Ntru~~!w Sc/w~ic&hrv~~~ Arch. P/~crrv~c~. 303: I33 13X. Bigalke, H., Hellcr, 1.. Bizzini. B. and Habermann. E. (1981). Tetanus toxin and botulinum A toxin inhibit release and uptake of various transmitters, as studied with particulate preparations from rat brain and spinal 4rch. Phtrrrwc~. 316: cord. Ntru,l!n-Sch,tli~~~b~~~/,s 244-251. Brooks. V. B., Curtis. D. R. and Eccles. J. C‘. (19%). The action of tetanus toxin on the inhIbItIon of motoneurones. J. P/~~siol. 135: 6.55 672. Collingridge, G. L.. Collins, G. G. S.. Davies. J., James. T. A., Neal, M. J. and Tongroach, P. (1980). Effect of tetanus toxin on transmitter release from the substantia nigra and striatum in ritro. J. Nrurochw1. 34: 54&547. Collingridge, G. L. and Davies, J. (1979). Tetanus toxin induced turning behaviour: A reversible reduction in GABA mediated transmission in the substantia nigra. Newosci. LL’US. suppl. 3: S222. Collingridge, G. L. and Davies. J. (1980a). Tetanus toxin inhibits I”HlGABA release from rat substantia nipra slices in ~~itro~ J. Physiol.. Land. 308: 72-73. Collingridge, G. L. and Davies, J. (1980b). Reversible effects of low doses of tetanus toxin on synaptic inhibition in the substantia nigra and turning behaviour in the rat. Brclirl Rcs. 185: 4.55%459.
Tetanus
toxin and GABA
Collingridge, G. L. and Davies, J. (1981). The influence of striatal stimulation and putative neurotransmitters on identified neurones in the rat substantia nigra. Brain Rrs. 212: 345-359. Collingridge, G. L., Thompson, P. A., Davies, J. and Mellanby, J. (1981). In vitro effect of tetanus toxin on GABA release from rat hippocampal slices. J. Neurochem. 37: 1039-1041. Curtis, D. R. and de Groat. W. C. (1968). Tetanus toxin and spinal inhibition. Brain Res. 10: 208-212. Davies, J. and Tongroach. P. (1979). Tetanus toxin and synaptic inhibition in the substantia nigra and striatum of the rat. J. Phq’siol., Land. 298: 23-36. Gale. K. and ladatola, M. J. (1980). GABAergic denervation of rat substantia nigra: functional and pharmaco_ logical properties. Brain kes. 183: 217-223. Gushchin, 1. S., Kozhechkin, S. N. and Sverdlov. Y. S. (1969). On the presynaptic nature of the suppression of postsynaptic inhibition by tetanus toxin. Boki. iikrrd. N&c, U.S.S.R. 187: 685-688.
release
855
Habermann, E.. Dreyer, F. and Bigalke. H. (1980). Tetanus toxin blocks the neuromuscular transmission in cirro like botulinunm-A toxin. ~u~~~~-Sc~~~~~~~~~gs Arch. Pharmat. 331: 33-40. Mellanby, J. and Green, J. (1981). How does tetanus toxin act. Neuroscirnce 6: 281-300. Osborne, R. H., Bradford. H. F. and Jones, D. G. (1973). Patterns of amino acid release from nerve-endings isolated from spinal cord and medulla. J. Nrurochm. 21: 407-4 19. Sherrington, C. S. (1906). The ~nr~~~~~till~ Action ofthe Ncr~0~~sS~srmm. Yale Uni~fersity Press. Simpson, L. L. (1980). Kinetic studies on the interaction between botulinurn toxin type A and the cholinergic neuromuscular junction. J. Phurmuc. up. T/w. 212: 16-21. Srinivasan, V.. Neal, M. J. and Mitchell, J. F. (1969). The effect of electrical stimulation and high potassium concentrations on the etflux of 3H-:-aminobutyric acid from brain slices. J. ,~~,~~r~~~~~~~7. 16: 1235-1244.
~28-3908/82~090851-05~3.~~0 Copyright 0 1982 Pergamon Press
Printed in Great Britain. All rights reserved
Ltd
THE IN VW-R0 INHIBITION OF GABA RELEASE BY TETANUS TOXIN G. L.
COLLINGRIDGE*
and J.
DAVIES
Department of Pharmacology, The School of Pharmacy, University of London, 29/39 Brunswick Square, London WCtN IAX, U.K. (Accepted 23 February 1982) in vitro effects of tetanus toxin on the potassium(K+)-evoked release of radiolabelled ;a-aminobutyric acid (GABA) from slices prepared from substantia nigra of the rat have been investigated. The addition of tetanus toxin (75(rlS,OOO mouse LD,,ml-‘) to the slices produced a dose- and time-dependent reduction in the K+-evoked release of C3H]GABA. These effects of the toxin were diminished at low tem~rature and in the absence of calcium ions in the bathing medium. It is suggested that this relatively simple in &ro system may allow more quantitative studies to be made concerning the action of the toxin on inhibitory amino acid release from the mammalian central nervous system.
Summary-The
Thousands of people are affected annually by tetanus and there is a high mortality rate since the mechanism of action of tetanus toxin is unknown and hence no specific treatment exists. The major locus of action of the toxin appears to be on the central nervous system (Brooks, Curtis and Eccles, 1957; Sherrington, 1906) where it has been shown to interfere with amino acidmediated synaptic inhibition by reducing the release of the transmitters, ;J-aminobutyric acid (GABA) and glycine, from nerve terminals (Collingridge, Collins, Davies, James, Neal and Tongroach, 1980; Curtis and de Groat, 1968; Davies and Tongroach, 1979; Guschin, Kozhechkin and Sverdlov, 1969: Osborne. Bradford and Jones, 1973). In these studies effects were achieved using either systemic or direct injection of the toxin into the whole animal. More information regarding the specificity and mechanism of action of the toxin in the CNS would probably be obtained if the toxin could be shown to be active following in vitro administration (cf experiments with botuIinum toxin, e.g. Simpson, 1980). Recently it has been reported in a preliminary form that the release of GABA from brain slices can be reduced by in citro administration of tetanus toxin (Colhngridge and Davies, 1980a; Collingridge, Thompson, Davies and Mellanby, 1981). Here the in vitro effects of tetanus toxin on C3H]GABA release from slices of rat substantia nigra are described in more detail.
were incubated in 2 ml oxygenated Krebs-Ringer bicarbonate solution (pH 7.4) at 37°C for an incubation period of 220min. Tetanus toxin (750-15,000 mouse LD5eml- ‘) was present for various periods (see Table legends) and C3H]GABA (8.7 x lo-‘M; 1 &i) was added for the last 30min of the incubation period. Slices were then transferred to small chambers and su~rfused with normal Krebs-Ringer bicarbonate solution at 37°C. After 24 min, consecutive 2 min fractions (2 ml) of the superfusate were collected, and at 30 min, release was evoked by increasing the K+ concentration of the Krebs-Ringer bicarbonate solution by 25 mM for 2 min. Aminooxyacetic acid (100pM) was present throughout to prevent metaboIism of C3H]GABA (Srinivasan, Neal and Mitchell, 1969) and the radioactivity released in each fraction was expressed as a percentage of tissue stores at that time. Evoked release was determined as the total release during and up to 10min after high K’ concentration minus the expected resting release at that time, which was calculated from the 8 min period preceding high K+ concentration, The release of C3H]GABA from slices incubated with toxin was compared, in parallel experiments, with the release from slices incubated under identical conditions in the absence of toxin or with tetanus toxin previously neutralized with tetanus antitoxin. RESULTS
METHODS The experimental methods have been described in detail elsewhere (Coilingridge er al., 1980). Briefly, 0.35 mm thick slices of substantia nigra from the rat * Present address for correspondence: Department of Physiology, The University of British Columbia, Vancouver, B.C., Canada V6T 1W5. Key words: tetanus toxin, substantia nigra, GABA, release.
Tetanus toxin had no effect on the spontaneous release of E3H]GABA from nigral slices, under any conditions employed. For example, following 210 min incubations in normal Krebs-Ringer bicarbonate solution, media containing active toxin (15,000 LD,e) or media containing neutralized toxin, the respective resting releases of 0.080 j, 0.004 (n = 25), 0.079 &- 0.009 (4) and 0.073 k 0.004 (4) “/, tissue stores min-’ were not significantIy different. Under appro-
852
G. L. COLLINGRIDGE and J. DAVIES
B
1
1 24
25 mM
1
I
26
-
I
k’
I
I
36
40
,
25 mM Time,
min
Fig. 1. Effect of tetanus toxin on the release of [jH]GABA from nigral slices. Each point is the mean + SEM of the release from four superfusion chambers containing slices (4-6 per chamber) exposed to tetanus toxin (dashed line) or control slices (continuous line). All eight superfusion chambers were run simultaneously. In (A) slices were incubated with 5000 mouse LD,, tetanus toxin or in the absence of toxin for 210 min before release experiments were commenced. In (B) slices were incubated for 210 min with 15,000 mouse LDso tetanus toxin or neutralized toxin (control).
priate conditions. however, active but not neutralized tetanus toxin, caused a significant reduction in the K+-evoked release of c3H]GABA from nigral slices (Fig. I). The effects of varying doses and exposure times of the slices to tetanus toxin on the release of C3H]GABA are summarised in Table I. Thus, with 210 min exTable 1. Effect of dose and exposure time with tetanus toxin on the K’-evoked release of C3H]GABA from nigral slices Interval of exposure to tetanus toxin (min) lc-220 lo-220 l&220 l&220 lo-40 l&70 16&220
Dose (mouse LDs, ml 750 2000 5000 15.000 5000 5000 5000
I)
Percentage reduction in K+-evoked release 12.2 25.1 51.7 66.0 7.7 45.9 8.8
k k + * k F +
7.4(6) NS 6.4(11)** 2.6(26)*** 3.7 (4) *** 8.5 (4) NS 9.2 (5)* 8.8(6)NS
Nigral slices were incubated for 220 min in Krebs solution. Tetanus toxin was present for the time shown in the left hand column. In experiments where toxin was added to the incubation media after 10 min and left in contact with the tissue for only 30 or 60min. slices were washed and incubations continued in normal Krebs solution for the remaining 180 and 150 min, respectively. Results are expressed as the mean + SEM percentage reduction in K+-evoked C3H]GABA release compared to the release in parallel control experiments. The figures in parenthesis refer to the number of pairs of superfusion chambers. Significance of difference (unpaired two-tailed l-tests) * P < 0.02: ** P < 0.01: *** P < 0.001.
posure to the toxin the effects were dose-dependent over the range examined (75&l 5,000 mouse LDsOmlm ‘). When toxin was present for the 10-70 min portion of the total 220min incubation, similar effects on release were seen, suggesting that the toxin had been irreversibly accumulated by the tissue within I hr. However. no significant effect was observed when the toxin was present for the 16&220 min period (Table 1) suggesting that an interval of time (greater than 30 min) must elapse after the toxin has been accumulated before appreciable effects develop. These results prompted a further experiment to investigate the time course of the effects of the toxin. Thus, slices were incubated in the presence or absence of tetanus toxin (5000 LDSo) for I hr before being washed and superfused for an additional 210min. Release was evoked 30 min after the start of superfusion in the normal manner, and at 60 min intervals thereafter: Figure 2 shows that 30 min after the slices had been removed from the toxin there was only a small reduction in K+-evoked GABA release (7.8”~ whereas 1 hr later there was a much greater effect (48.3”,, reduction). At this time the effects of the toxin appeared almost maximal since similar reductions were observed I and 2 hr later. Finally. preliminary studies were performed to investigate the effect of incubation temperature and of Ca2+ ions on the action of the toxin. When slices were incubated with tetanus toxin for ISOmin at 2.5 C and then at 37’ C for 30 min (while loading the slices with C3H]GABA) instead of at 37 ‘C throughout
853
Tetanus toxin and GABA release 0.B
r
Time,
min
Fig. 2. Time course of the effects of tetanus toxin on GABA release. Slices were incubated for 1 hr in the presence or absence of tetanus toxin (5000 mouse LDso) and then superfused for 222min. C3H]GABA release was evoked by 2 min K+ pulses (25 mm) at 30 min and hourly intervals thereafter. The 4 resulting release peaks are illustrated for control (continuous) and toxin-incubated slices (dashed line), but 40 min intervals between K’ pulses have been omitted. See text for further discussion of Figure. This experiment was performed twice and yielded similar results both times.
there was no change in the accumulation of C3H]GABA or its subsequent spontaneous release, However, there was a considerable reduction in the K+-evoked release (-70%; release performed at 37°C). When compared to parallel controls, the toxininduced reduction of K+-evoked release was smaller following incubation at low temperature (Table 2) suggesting that the action of the toxin may be temperature-dependent. Omitting Ca*+ ions from the media for the total 220 min incubation period did not significantly affect the uptake, spontaneous or evoked release of C3H]GABA (release being performed in the presence of Ca* ‘) but did reduce the effect of the toxin on the evoked release of C3H]GABA from nigral slices (Table 2).
The effects of tetanus toxin on K+-evoked C3H]GABA release cannot be attributed to a reduction in uptake since release is expressed as a percentage of tissue stores. Additional effects on uptake seem unlikely since the radioactivity remanning at the end of the experiment in control and toxin-incubated slices was never significantly different. (There was often slightly more radioactivity remaining in the toxinincubated slices which can be attributed to the smaller amount released from these slices). Furthermore, in separate uptake experiments no effects of the toxin were detected. For example, following incubation for 180 min with 2000 mouse LD,, tetanus toxin or in the absence of toxin the respective IO min tissue: medium ratios (dpm g- 1 tissue: dpm ml- I
Table 2. Effect of temperature and Ca ‘+ ions on the action of tetanus toxin
Temperature (“C) 37 2.5 37 37
Ca2+ ion cont. (mM) 2.52 2.52 2.52 0
K+-evoked release [‘HIGABA (“/, tissue stores) Control 5.77 + 1.69 + 4.13 5 4.90 f
0.79 0.13 0.33 0.54
Toxin
Percentage reduction in K “-evoked release of cJH]GABA
2.25 & 0.39 1.17 + 0.59 1.85 &-0.19 2.85 + 0.17
61.0 _t 6.7 (4) 30.6 k 3.5 (4)* 55.2 t 4.7 (6) 41.9 + 3.4(6)**
Nigral slices were incubated for 210 min with tetanus toxin (5000 mouse LD,, ml- ‘) and the mean k SEM release from these slices (fourth column) was compared with the release in parallel control experiments (third column). The fifth column presents the mean k SEM% reduction compared to the mean control value and the figures in parenthesis refer to the number of pairs of superfusion chambers. In the first series of experiments the release from slices (control and toxin) incubated for 180 min at 2.5”C and then 30min at 37°C was compared with slices incubated for 2lOmin at 37°C. In the second series, the effects of omitting Ca ‘+ ions from the incubation medium for 210 min was examined. Significance of difference in toxin-induced reduction in K+-evoked release. *P < 0.05;**p < 0.001.
G.
854
L.
COLLINGRIDGE and J. DAVIES
medium; n = 4) of 6.9 k 0.8 and 7.7 k 0.8 were not significantly different. DISCUSSION
In previous studies it was reported that tetanus toxin reduced Ca”-dependent K’-evoked release but not uptake, the small Ca”-independent K+-evoked release or the spontaneous release of 13H]GABA from nigral slices following in L’I’UO pretreatment with the toxin (Collingridge et al.. 1980). The present results confirm that tetanus toxin reduces the K+-evoked release of C3H]GABA from nigral slices in a Ca’ +-containing medium. It is likely, therefore, that the reduction in GABA-mediated synaptic inhibition in the nigra seen in ciw following intranigral injections of the toxin (Collingridge and Davies, 1980b, 198 1; Davies and Tongroach, 1979) is due to an effect on the synaptic release of GABA. The novel observation is that C3H]GABA release could be affected by in citro administration of the toxin under appropriate conditions, implying that the nigral slice contains all the necessary structures for tetanus toxin to act. Since in this slice preparation GABAergic terminals are largely or totally severed from their cell bodies (Gale and Iadarola, 1980) and since the K+-evoked release of C3H]GABA is probably from nerve terminals (Collingridge et N/., 1980). the toxin presumably affects release via an action directly on nerve terminals rather than indirectly via an effect on axons or cell bodies. The effects of tetanus toxin developed slowly. Thus, over 90min was required, after first exposure of the slices to the toxin, before a significant reduction in C3H]GABA release was detected (Table 1) and maximal effects took approx. 150min to develop (Fig. 2). This slow time course of the action of the toxin probably explains why no effect was observed on the electrically-evoked release of endogenous GABA or glycine only 30 min after in vitro administration of the toxin to synaptosomes from the rat brain (Osborne cut (II., 1973). Recently, Bigalke. Heller, Bizzini and Habermann (1981) have also reported an in vitro action of tetanus toxin on GABA release from the CNS. They observed similar reductions in [‘HIGABA release, to those reported here. following 2 hr incubations of rat brain particles with doses of toxin comparable to those used in the present study. Interestingly, in the present study it was only necessary for the toxin to be present for a relatively short period of time (greater than 30 min but less than 60 min) for full effects to develop, providing incubation of the slices was then continued in the absence of toxin. This observation suggests that the toxin is irreversibly accumulated by the slices within 1 hr and then an additional slow process is necessary before release is affected. The 3&60 min exposure time of the slices necessary for effects to develop subsequently was probably determined by the slow diffusion of the toxin molecules through the slices rather than the
actual binding process which is believed to be much more rapid. The effects of tetanus toxin appear not to be specific towards the release of inhibitory amino acid transmitters since the release of dopamine was reduced from slices prepared from toxin-poisoned striata (Collingridge et al.. 1980) and acetylcholine and noradrenaline release were decreased when brain particles were incubated with the toxin (Bigalke c’f I//.. 1981). Furthermore, acetylcholine release from primary nerve cell cultures (Bigalke, Dimpfel and Habermann. 1978) and from peripheral preparations (Habermann. Dreyer and Bigalke. 1980) was reduced following incubation with tetanus toxin. However, the release of GABA seems to be particularly sensitive to the action of tetanus toxin (Bigalke rt rrl.. 1981; Collingridge clr crl., 1980) and in the nigra is reduced by pretreatment with sublethal doses of the toxin (Collingridge and Davies. 1979). These and other data (Mellanby and Green. 1981) suggest that effects on GABA release are involved in tetanus. Therefore, this preparation should be useful in elucidating the pathogenic mode of action of the toxin and in the evaluation of drugs for specific antitetanus properties. Finally. this relatively simple system should enable quantitative studies to be made concerning the actions of the toxin on GABA release analogous to those performed with botuiinum toxin on acetylcholine release at the neuromuscular junction (Simpson, 1980). Ac,k~~o~~/rdyt,mr,nr,s~ We thank Dr M. Simmonds ing the manuscript Davey for technical
and Rachel assistance.
Williamson
and
for readRichard
KEFERENCES
Bigalke, H.. Dlmpfel. W. and Habermann. E. (1978). Suppression of ‘H-acetylcholine release from primary nerve cell cultures by tetanus and botulinurn-A toxin. Ntru~~!w Sc/w~ic&hrv~~~ Arch. P/~crrv~c~. 303: I33 13X. Bigalke, H., Hellcr, 1.. Bizzini. B. and Habermann. E. (1981). Tetanus toxin and botulinum A toxin inhibit release and uptake of various transmitters, as studied with particulate preparations from rat brain and spinal 4rch. Phtrrrwc~. 316: cord. Ntru,l!n-Sch,tli~~~b~~~/,s 244-251. Brooks. V. B., Curtis. D. R. and Eccles. J. C‘. (19%). The action of tetanus toxin on the inhIbItIon of motoneurones. J. P/~~siol. 135: 6.55 672. Collingridge, G. L.. Collins, G. G. S.. Davies. J., James. T. A., Neal, M. J. and Tongroach, P. (1980). Effect of tetanus toxin on transmitter release from the substantia nigra and striatum in ritro. J. Nrurochw1. 34: 54&547. Collingridge, G. L. and Davies, J. (1979). Tetanus toxin induced turning behaviour: A reversible reduction in GABA mediated transmission in the substantia nigra. Newosci. LL’US. suppl. 3: S222. Collingridge, G. L. and Davies. J. (1980a). Tetanus toxin inhibits I”HlGABA release from rat substantia nipra slices in ~~itro~ J. Physiol.. Land. 308: 72-73. Collingridge, G. L. and Davies, J. (1980b). Reversible effects of low doses of tetanus toxin on synaptic inhibition in the substantia nigra and turning behaviour in the rat. Brclirl Rcs. 185: 4.55%459.
Tetanus
toxin and GABA
Collingridge, G. L. and Davies, J. (1981). The influence of striatal stimulation and putative neurotransmitters on identified neurones in the rat substantia nigra. Brain Rrs. 212: 345-359. Collingridge, G. L., Thompson, P. A., Davies, J. and Mellanby, J. (1981). In vitro effect of tetanus toxin on GABA release from rat hippocampal slices. J. Neurochem. 37: 1039-1041. Curtis, D. R. and de Groat. W. C. (1968). Tetanus toxin and spinal inhibition. Brain Res. 10: 208-212. Davies, J. and Tongroach. P. (1979). Tetanus toxin and synaptic inhibition in the substantia nigra and striatum of the rat. J. Phq’siol., Land. 298: 23-36. Gale. K. and ladatola, M. J. (1980). GABAergic denervation of rat substantia nigra: functional and pharmaco_ logical properties. Brain kes. 183: 217-223. Gushchin, 1. S., Kozhechkin, S. N. and Sverdlov. Y. S. (1969). On the presynaptic nature of the suppression of postsynaptic inhibition by tetanus toxin. Boki. iikrrd. N&c, U.S.S.R. 187: 685-688.
release
855
Habermann, E.. Dreyer, F. and Bigalke. H. (1980). Tetanus toxin blocks the neuromuscular transmission in cirro like botulinunm-A toxin. ~u~~~~-Sc~~~~~~~~~gs Arch. Pharmat. 331: 33-40. Mellanby, J. and Green, J. (1981). How does tetanus toxin act. Neuroscirnce 6: 281-300. Osborne, R. H., Bradford. H. F. and Jones, D. G. (1973). Patterns of amino acid release from nerve-endings isolated from spinal cord and medulla. J. Nrurochm. 21: 407-4 19. Sherrington, C. S. (1906). The ~nr~~~~~till~ Action ofthe Ncr~0~~sS~srmm. Yale Uni~fersity Press. Simpson, L. L. (1980). Kinetic studies on the interaction between botulinurn toxin type A and the cholinergic neuromuscular junction. J. Phurmuc. up. T/w. 212: 16-21. Srinivasan, V.. Neal, M. J. and Mitchell, J. F. (1969). The effect of electrical stimulation and high potassium concentrations on the etflux of 3H-:-aminobutyric acid from brain slices. J. ,~~,~~r~~~~~~~7. 16: 1235-1244.