Ex vivo release of GABA from tetanus toxin-induced chronic epileptic foci decreased during the active seizure phase

Neuro~hem lnt Vol 18, No 3, pp 373 379, 1991 Printed m Great Britain

0197-0186/91 $3 00+0 00 Pergamon Press plc

E X VIVO RELEASE OF GABA FROM TETANUS TOXIN-INDUCED CHRONIC EPILEPTIC FOCI DECREASED D U R I N G THE ACTIVE SEIZURE PHASE JOHN G. R. JEFFERYS, t PHILIP MITCHELL,~ LAWRENCE O'HARA, J CHRISTOPHER TILEY, ~ JOHN HARDY, 2 SARAH J JORDAN, i MARINA LYNCH 4 a n d JANE WADSWORTH3 Departments of ~Physiology and Biophysics, 2Biochemistry and Molecular Genetics and 3pubhc Health, St Mary's Hospital Medical School, Imperial College of Science, Technology and Medicine, London W2 1PG, U K 4National Institute for Medical Research, Mill Hdl, London NW7 IAA, U K ( Recewed 6 June 1990, accepted28 August 1990) Al~raet--Injectmg a few mouse LDs0 of tetanus toxin into rat hlppocampus has been shown to induce a remarkably persistent sequence of functional changes which provide a chromc model of limblc epdepsy. Here we have measured the release of amino acid transmitters evoked by K+-stxmulation from hlppocampal shces prepared from rats which had been injected 10-14 days previously with 6 mouse LDs0 (c. 3 ng) of tetanus toxin The Ca2+-dependent component of the release of [14C]y-ammobutync acid (GABA) was depressed to two thirds its control level Rats which had survived 6-8 weeks, by which time the seizures had ceased, showed a recovery of the Ca-'+-dependent component of the K÷-evoked release of GABA to control levels, but these rats also exhibited a paradoxical depression of the Ca2+-independent component of release [3H]D-Aspartate has previously been used as a putatwe marker for excitatory amino acid release However, it failed to fulfil this role m the present study because its release was not stimulated by K + In contrast [3H]D-aspartate was released in response to veratrme Together with previous work this suggests that while [3H]D-aspartate was taken up into neurones, it did not enter the releasable vesicular pool HPLC measurements of the release of endogenous excitatory amino acids showed that glutamate (and not aspartate) was stimulated by K + m a Ca2+-dependent manner, and that the amount of release did not differ in the tetanus toxin-injected rats The depression of GABA release provides the most hkely mechanism for the seizures The recovery of its Ca2+-dependent release provides the most hkely basis for seizure remission after 6-8 weeks, m this chronic epileptic syndrome

T e t a n u s toxin is a p o t e n t n e u r o t o x l n which blocks synaptlc transmission, a p p a r e n t l y by interfering with the coupling o f exocytosls with increased mtracellular calcium (Bevan a n d W e n d o n , 1984 ; Bittner a n d Holz, 1988). Previous studies o f the t o x m ' s acute effects in vitro o n b r a i n shces ( C o l h n g n d g e a n d Davies, 1982; Collingridge et a l , 1981) o n cultured n e u r o n e s (Bergey et al., 1983, H a b l g et al., 1986) a n d o n b r a i n h o m o g e n a t e s (Albus a n d H a b e r m a n n , 1983) have suggested t h a t ~t blocks the release o f G A B A , thus interfering with synaptlc inhibition, a n d p r o v i d i n g a basis for chmcal m a m f e s t a t l o n o f tetanus. However, the release o f o t h e r n e u r o t r a n s m i t t e r s is also affected by the toxin (Albus a n d H a b e r m a n n , 1983, Col-

hngridge et al., 1980), a l t h o u g h there m a y be some selectivity for G A B A (CollIngndge et a l , 1980), which IS d e p e n d e n t b o t h o n the dose o f toxin a n d the duration o f it application ( H a b i g et a l , 1986). W e have been particularly interested in the long term effects o f injecting a few m o u s e LDs0 o f tetanus toxin into rat h i p p o c a m p u s Th~s induces a n epileptic s y n d r o m e which provides a m o d e l o f h u m a n h m b i c epilepsy (Brace et al., 1985; Jefferys a n d Wilhams, 1987; H a w k i n s a n d Mellanby, 1987, M e l l a n b y et al., 1977, 1984, Jefferys, 1989). There are two phases to th~s s y n d r o m e First, p a r t m l a n d secondarily generalized seizures recur intermittently for a period o f several weeks starting a few days after injection. Second, after 6-8 weeks all signs of seizure actwaty cease, b u t the rats retain a p p a r e n t l y p e r m a n e n t func-

* Author to whom all correspondence should be addressed 373

~74

IIIHN (a R JrH4R',S el a/

tlonal a b n o r m a h t m s which include behavloural lmpmrments, e g of learning and memory (Mellanby et a l , 1982), a n d t h e d e p r e s s m n o f h l p p o c a m p a l c v o k e d r e s p o n s e s (Brace et a l , 1985, Jefferys a n d W f l h a m s , 1987) A t n o s t a g e m this m o d e l is t h e r e e v i d e n c e o f cell loss ( M e l l a n b y et a l , 1977, Jefferys et a l , 1988), a l t h o u g h o t h e r s t u d i e s h a v e s h o w n s o m e loss o f p y r a m i d a l cells after t h e l n j e c t m n o f m u c h h i g h e r d o s e s (c 100 500 t i m e s ) o f t e t a n u s t o x i n ( B a g e t t a et a l , 1990) Acute experlinents suggest that tmpmred GABA release m a y be r e s p o n s i b l e for t h e c h r o m c e p l l e p t m s y n d r o m e r e d u c e d by t h e l n t r a h l p p o c a m p a l m j e c t m n ( C o l h n g r l d g e et a l , 1981) T h e s e a c u t e s t u d m s u s e d 250 m o u s e LD~() a p p h e d , m 2 ml o f a m f i c i a l cere b r o s p m a l fired (a c s f ) , to 20 h l p p o c a m p a l slices (c. 60 m g ) W h i l e it Is h a r d to c o m p a r e this & r e c t l y to t h e d o s e s a p p h e d m t w o , it ts g r e a t e r t h a n t h e 6 LD~o rejected to r e d u c e t h e c h r o m c s y n d r o m e . I n d e e d , s t u d lCS o f t h e fate o f r a & o l a b e l l e d t o x i n injected i n t o t h e h l p p o c a m p u s s h o w e d t h a t 1 5 % o f t h e original d o s e w a s left a f t e r 9 d a y s ( M e l l a n b y , 1989) H e r e we h a v e s t u d m d n e u r o t r a n s m ~ t t e r release in t h e c h r o m c s y n d r o m e , u s i n g p o t a s s m m - e v o k e d release o f a m m o a c i d s f r o m b r a i n s h c e s p r e p a r e d f r o m rats injected w i t h the t o x m s o m e d a y s to w e e k s p r e v i o u s l y S h c e s were p r e p a r e d 10-14 d a y s a f t e r reJection to e x a m i n e t h e active s m z u r e p h a s e o f t h e s y n d r o m e , a n d 6- 8 w e e k s after l n j e c t m n , to s t u d y t h e m e c h a n i s m s u n d e r l y i n g lhe remission from smzures

EXPERIMENTAL PROCEDURES

Sprague Dawley rats weighing 280 320 g (Harlen O L A C Ltd, BlCester, U K ) were anaesthetized with pentobarNtone (Sagatal. 60 mg,,kg, May & Baker), and received 1 ~1 injections of e,ther tetanus toxin (gift of Wellcome Blotech, Beckenham. U K ) or vehicle solution (0 1 M phosphate buffer containing 0 2% gelatin) into both hlppocampl from a Ham~lton syringe mounted on the K o p f stereotaxlc frame Each toxin injection contained 3 ng of protein, or about 6 mouse LDs~, The rejections were made into the dorsal hlppocampus at coordinates 2 5 m m caudal to bregma, + 3 5 m m lateral and t l m m below the surface of the cortex (Pellegrlno et ul, 1979) For further details of the methods see (Brace et al, 1985, Jefferys and Williams, 1987) The rats were allowed to recover and were housed 2 or 3 to a cage Transmitter release was measured using shces of the dorsal hlppocampl of both sides, prepared either 10 14 days or 6 8 weeks after injection Each rat rejected with tetanus toxin was paired with one which had received a control rejection The rats were killed by stunning and cervical dislocation, their brains were rapidly dissected free and 300 #nr transverse (parasagltal) hlppocampal slices were cut on a Vlbroshce (Campden Instruments, London) Eight slices from each animal were transferred to a chamber containing 2 ml of artificial cerebrosplnal fluid

(a c s f ) This solution, like all the others used, was freshly equilibrated with a 95% O_~, 5% CO~ gas m~xture and was maintained at 37 C It contained 135 m M NaCI, 3 m M KCI, 16 m M N a H C O , , 1 25 m M NaH2PO4, I m M MgCI,, 2 m M CaCI> 10 m M glucose, 100 #M amlno-oxyacetlc amd (to prevent the metabohsm of GABA) and 1 #M tetrodotoxln [to stop spontaneous epileptic activity whmh is preserved m slices from these chronically epileptic rats (Jefferys, 1986, 1989)] After 10 mm, the solution was exchanged with one containing the two radmlabelled compunds, 25 itCl [~4C]GABA and 2 5 ,uC~ [~H]D-aspartate (both from Amersham), and incubated for 15 rain After loading with the radlolabel, the shces were washed 3 t~mes with fresh, unlabelled a c s f and transferred to ,in array ol 8 water-jacketed chambers, whmh were perfused with a c s f at about 0 45 lnl,'mln Four chambers contained slices from the toxin reJected rat, and 4 from the control Alternate chambers were perfused throughout the experiment with solutions contaming 0 m M Ca -'~ and 3 m M Mg -~+, the remainder V,lth 2 m M Ca 2' and 1 m M Mg -'~ The lirsl 15 rain of perfusate ~ a s discarded to wash off excess label Then, a total of 22 consecutive 2 lnln samples of perfusate ~ere collected Between 15 and 35 mln after the start of the run, the perfusate was changed to one with 30 m M KC1 instead of 3 m M NaCI concentratmns being reduced to maintain osmolant> At the end of the experiment the shoes were collected and d~srupted in 1 ml of water Four ml of Ecosclnt was added to each sample v, hlch were counted on a Packard T n c a r b scintillation counter The raw counts ~ere transformed to express the release of each radlolabel as a fraction of the a m o u n t ol label left m the tissue (using a spreadsheet faclhty in the RSI software package running under MS DOS) The experiments to measure the release of endogenous amino acids used duplicate 3 mm samples collected at 9 and 12 m m Into the experiment for "'pre-stlmulatlon'" release, at 21 and 24 m,n lor "'stimulated" release, and at 42 and 45 mln for "'post-stlmulatmn'" release (the 30 m M K + again being applied from 15 to 35 Iron) These samples were freezedried and stored The samples were mixed with I ml of methanol/,~,lter (1 I, ~,~), vortex mixed and kept at 37 (' for 30 m m The> were then vortex mixed again, centrifuged at 1000 g for 2 m m and 4011 td of the supernatanl taken for analysis using high pert\~rmance llqmd chromatography (HPLC) Ahquots ( 100 ltl) of these supernatant samples were transferred mto a mixing chamber with an equal ~olume of o-phthalaldehyde,'mercaptoproplomc acid reagent by an autosampler (MlcromerltlCs 725 automjector), and 100 l~l ahquots of this were injected after 1 mm of mixing onto the column (Mlcrosorb C18, 10 cm) The separated amino acid derivatives were detected by fluorescence at 360 470 nln with a Kratos 950 fluonmeter Analysis of variance was earned out using the SAS package at the Umverslty of London C o m p u t m g Centre in order to assess the statistical significance of the findings

RESULIS Eplh'ptu ~ y m h o m e T h e rats injected with t h e t e t a n u s t o x i n closely r e s e m b l e d t h o s e we h a v e d e s c r i b e d p r e v i o u s l y (Brace el a ] , 1985. lefferys a n d W l l h a l n s 1987) Visual observ a t i o n o f t h e rats c o n f i r m e d t h a t t h e y h a d b e c o m e

375

Ex vwo release of GABA in the active seizure phase epileptic by 3-7 days after injection T h e potency o f the toxin was f u r t h e r confirmed by electrophyslologlcal studies of the eplleptiform discharges which were f o u n d reliably in h l p p o c a m p a l slices prepared from o t h e r rats Injected with the same b a t c h o f toxin (Jefferys, 1989) G A B A release durmg set,cure phase T h e first part of the experiment used rats which h a d survived 10-14 days after the injection o f tetanus toxin or vehicle solution (n = 7 in each case), 1.e d u n n g the active seizure phase The baseline fractional release o f [' 4 C ] G A B A was u n i f o r m a n d no differences were detected between slices p r e p a r e d from tetanus toxin a n d control injected rats, n o r between those exposed to a c . s . f c o n t a i n i n g 0 a n d 2 m M Ca 2+ [Fig. I(A)]. K +-stImulatIon (30 m M ) of the slices caused a m a r k e d increase in the release o f []4C]GABA, after a delay o f 3 5 m m which can be attributed to the dead space in the perfuslon system This increase was sustained for a b o u t 10 rain before it started to decrease. T h e release a p p r o a c h e d the prestlmulus level by the end o f the experiment, some 10 mln after the K + h a d been restored to 3 m M The K + - s t l m u l a t e d release was calculated for each tube by s u b t r a c t i n g the m e a n value o f the unstlmulated release between 6 a n d 14 rain into the r u n from that for each value o b t a i n e d between 22 a n d 28 rain

002

(thus there were 4 replicates for each o f the 4 slice c h a m b e r s used per rat). There was a significant Ca 2+ dependence of the stimulated [' 4 C ] G A B A release, as would be expected for release from the synaptIc transmitter pool, a n d there was a significant difference between the tetanus toxin a n d control injected groups ( P < 0.0001 for b o t h factors, 2-way A N O V A with interactions). M o s t i m p o r t a n t l y , there was a significant interaction between these two factors ( P = 0.0036), which suggests t h a t the toxin injected g r o u p h a d a different (lower) level o f K + - s t i m u l a t e d , C a 2 + - d e p e n d e n t release o f []4C]GABA [Fig I(B)] There was n o sign o f any effect o f the toxin injection o n the C a 2 + - l n d e p e n d e n t stimulated release G A B A release during the setzure remission phase F r o m the extensive observations o n this model we can be confident t h a t electrographlc seizures a n d m o t o r fits would have ceased c 6 weeks after injection o f the toxin : visual observations o f the rats before the start o f these measurements, together with parallel electrophyslologlcal studies o n slices from o t h e r rats (S J. J o r d a n a n d J. G. R Jefferys, u n p u b l i s h e d observations) gave us n o reason to d o u b t t h a t the rats h a d indeed g a m e d a remission from their seizures T h e basehne release o f [ ' 4 C ] G A B A was uniform a n d there were n o significant differences between the experimental groups [(n = 5 for e a c h , Fig. 2(A)] The

(A)

+

"-'/ "\'~, 001

0 020

~

oo1~

o

0010

o c

0005

(B)

0

o

o U-

I

10

210 Time (rnln)

30

[

40

o

I

I

I

0

2

0

Control group

I

I

2 mM Ca

Tetanus Grand group mean

Fig I Mean ['4C]GABA release rata the a c s f superfuslng slices prepared 10~14 days after rejection, 1 e during the period or active seizures Slices were taken from 7 control rats and incubated m the presence (@) and absence (O) of Ca 2+ , and from 7 tetanus toxin rejected rats with (A) and without (A) Ca -'+ (A) Fractional release of labelled toxin as a function of t]me, [K+]0 was increased from 3 to 30 mM between 15 and 35 rain (indicated by the horizontal bar), and clearly stimulated release (B) The mean Increase in release caused by the K + stimulation has been plotted for the control and toxin-rejected groups. in the presence and absence of Ca -'+ in the a c s f to summarize the two-way analys~s of variance [symbols as for (A), except • which is the overall, or "'grand mean" of the increase in release on increasing [K+]0, error bars are _+residual root mean square error]

X76

h)Hb, (J 0 04

R

JIIFFRYS ('1 U/

(A)

~

oo4

(B)

003 o

"~x,.

°

8 ?, 00z

-~ 0 0 2

h

•

T

-~ o 001

OO1

¢_o--o-o-¢-~-TJp

I

I0

.

.

2LO

.

.

.

310

T,me (m,n)

o * "~'t~

[

40

zx

o

I 0

i 2

ControL group

I 0

I I 2 mM Co

Tetanus group

Grand mean

Fig 2 Mean [3C]GABA releabe from slices prepared 6 8 weeks after mtrah~ppocampal mjecUons, 1 c following seizure remission See legend to Fig I for details

levels of the baseline release were rather greater t h a n ( a p p r o a c h i n g double) those f o u n d during the first part of the study [Fig I(A)] This could be due to uncontrolled variations in o u r techmque, ageing or e n v i r o n m e n t a l factors arising from the longer time spent by this group m our a m m a l house W h a t e v e r the cause of this d~fference, it u n d e r h n e s the xmportance of the tight pairing of experimental and control rats used during these m e a s u r e m e n t s There again was a Ca ~'+ dependence of the K +evoked release o f [14C]GABA, and there was also a sigmficant difference between the tetanus toxin a n d control groups ( P < 0 0001) However, in c o n t r a s t with the results for the active seizure phase, there was no s~gnificant interaction between the effects of Ca ~'~ and the type of injection ( t o x i n / c o n t r o l . P = 0 47) Thus it was the Ca2+-mdependent, K+-evoked release of [~4C]GABA which was depressed in the tetanus toxin-rejected rats (or e n h a n c e d in the control rats) The CaZ+-dependent c o m p o n e n t (1 e the difference between the total stimulated release and the Ca 2+independent c o m p o n e n t ) was similar in the two groups of rats [Fig 2(B)]

E,ccttalo~]' amino acids durm,q the actwe seizure phase Thc shccs wcrc loaded with [~H]D-aspartate at the samc time as the [t4C]GABA, in order to label the C×cltatory a m i n o acid t r a n s m i t t e r pool The relcase data showed that the D-aspartate was not a good marker for the CXCltatory transmJtter m these cxperiments (Fig 3) There was no evidence of an) K ~stimulation of the release, n o r of any effect of the prescnce of Ca z~ ( P > 0 2) The release was slightly

greater from slices from toxin-injected t h a n from vehicle-injected rats ( P = 0.002) We were concerned at the lack of K + stimulation of [~H]D-aspartate release and therefore repeated these experiments in the absence of tetrodotoxln a n d of amlno-oxyacetic acid K + still failed to evoke release. In c o n t r a s t veratrlne was able to elicit release (Fig 3(C)] These data support the c o n t e n t i o n (Nlcholls et a / , 1987, Sanchez-Prleto et u l , 1987) that exogenous D-aspartate does not enter the releasable vesicular pool, and therefore that radlolabelled o - a s p a r t a t e is not a suitable m a r k e r for glutamate as a neurot r a n s m i t t e r in the h l p p o c a m p u s In c o n t r a s t we have observed K + sumulated [~H]D-aspartatc release from rat strlatal slices We repeated these experiments m the absence of the radlolabelled c o m p o u n d s , on a n o t h e r group of 6 control and 6 toxin-rejected rats, using H P L C to measure the release ol endogenous a m i n o acids In this case we had to use fewer replicates (duplicates for each c o n d m o n ) These data showed that glutamate release was stimulated by 30 m M K ~ in a C a - " d e p e n d e n t m a n n e r ( P = 0 02), but aspartate was not (P = 0 46, Fig 4) The release of e n d o g e n o u s glut a m a t e was affected by the toxin rejection ( P = 0 02), but it was the CaZ+-lndependent, and not the Ca-" dependent component ( P = 0 6), which v~as depressed DI~,CI, S M O N

The central result of this study is that the K ~stimulated, C a ~ - d e p e n d e n t release of G A B A was

377

Ex vwo release of GABA m the active seizure phase 002

0

3

001

9

00o4 - ( B )

(A)

I

o

oooz

= o

_J

= o

o - 0.002

I

,

10

20 Time

0 05

. 30

) 40

h

-0004

(mm)

I

o 2 Control group

(C)

I 0

I I 2 mM ca

Tetanus group

Grand mean

004

oo3 om 0 0 2 t.) u_ 0 0 1

I 10

I

z;

3'0

40

5'o

T=me (ram)

Fig 3 Mean [3H]D-aspartate release from shces prepared 1(~14 days after mtrahlppocampal rejections (A, B) See legend to Fig. 1 for details, note the complete lack of st]mulatmn by 30 mM [K+]0 (C) Veratrlne caused a marked release of [3H]D-aspartate when added to the perfuslng a c s f, at the end of an expenment similar to that shown m (A) depressed during the period of recurring seizure activity following the lnjecUon of a minute dose of tetanus toxm into rat hlppocampus. This therefore supports the hypothesis that impaired GABA release was responsible for the seizures, which had been suggested by release stu&es using the acute apphcation of high doses of tetanus toxin (Colhngrldge et e l , 1981, Collingrldge and Davies, 1982). With longer survwal t~mes (6-8 weeks), the Ca z+dependent K+-evoked release of GABA was restored to control values in the tetanus toxin injected rats, which provides a plausible explanation for the seizure remission which invariably occurred at this stage of the syndrome (Brace et el., 1985 ; Jefferys and Williams, 1987, Hawkins and Mellanby, 1987, Mellanby et el., 1977) Both the seizure remission and the recovery of the CaZ+-dependent release of GABA may be attributable to the removal of the toxin from the Ussue, which has a time course of the order of several days (Mellanby et e l , 1984, Hablg et el., 1986, Mellanby, 1989) One of the remarkable features of the tetanus toxin epileptic syndrome is the persistence of abnormal

behaviour and neurophyslological function (Brace et e l , 1985, Jefferys and Williams, 1987, Mellanby et el., 1977) The present study also revealed a persistent, though unexpected, change in the Ca2+-lndependent release of GABA. This component of the release could be dependent on the very low levels of Ca 2+ which may remain present under nominally Ca2+-free conditions If this were the case, then the lower levels of release in the toxin-injected rats could then be due to a continued impairment of the coupling of Ca 2+ to exocytosls which was overcome by normal levels of C a 2+. We feel this explanaUon is unlikely because the presence of increased Mg 2+ would be expected to block Ca 2+ entry. Alternatively, fins release may come from a non-synaptic, cytoplasmic pool of GABA in neurones, or possibly glia (Szerb et e l , 1981) It is tempting to speculate that this persistent decrease in Ca2+-lndependent release may be related to the electrophyslologlcal hypoexcltabllity we have described previously at this stage of the syndrome (Brace et e l , 1985 ; Jefferys and Williams, 1987), but at present the mechamsm remains unclear The 1dee that inhibitory synapses may be compro-

378

JOHN ( J

JEFFERYS ~'[ (ll

R

400 - (A),

c

150

(A)u

300

•

rO E

~

200 --

10C

O.

1 Control

I StimuLated

I 0

I Recovery

~ 2

Control group 2`50

(B)i

Iso

•~

•

I 0

t L 2 mMCo

Tetanus Grand group

1O0

E .~

1,50

"' IO0

50

:

"':

"

E to

5O

-~

o

......... o ~ ~ I Control

I StimuLated

I Recovery

~-so -100

mean

(Bill

T

2OO E

!

o `50

" °

I I 0 2 Control group

. z~

! I

I I I 0 2 mM Co Tetanus Grand group mean

Fig 4 Release of endogenous excitatory amino acids rata the a c s t superfuslng shces prepared l0 14 days after mtrahlppocampal injections, as measured by HPLC (A)I, Mean release of glutamate before, during and after K + stimulation (symbols as in Fig 1 ) , . , mean increase in glutamate release snmulated by K + (B)L n, Corresponding data for aspartate, revealing much weaker (and non-significant) effects of K + stlmulahon

mlsed m expenmental and chmcal epilepsies has a long and mixed history (reviewed m Jefferys and Roberts, 1987) There is evidence that seizure activity can m itself impair lnhlbmon, a particularly relevant example being the transient depression of G A B A release from hippocampal slices following electroconvulslve shock or ECS (Green et a l , 1987). Curiously, kindling with more focal stimulation appears to have the opposite effect, enhancing G A B A release from the hlppocampal C A I region (Kamphuls et a l , 1990) The effects of ECS and tetanus toxin were in the same direction, but had totally different rime courses. Those for ECS could be detected between 4 and 30 mm after the last convulsion, depending on the number experienced (Green et a l , 1987, Green and Vincent, 1987). In the case of the tetanus toxin model, no motor seizures were noted during the period of up to 30 mln the rats were m the laboratory before the terminal experiment, and indeed few would be expected given their low frequency with the doses of toxin we use (Jefferys and Wdhams, 1987) Therefore we believe it

unlikely that the depression of release seen with the tetanus toxin model was due to the mechanisms involved in the short term effects of ECS The selectivity of the changes during the active seizure phase of the intrahlppocampal tetanus toxin syndrome is an important ~ssue given the range of transmitters which are affected by acute applications of high doses of the toxin Unfortunately, our results have exposed a limitation In the effectiveness of [3H]Daspartate as a marker for the release of excitatory amino acids, although it has previously proved effective In several preparations (Skrede and Malthe Sorenssen, 1981) It is possible that D-aspartate can mark only aspartate and not glutamate transmitter pools In parallel experiments without radiolabel, we observed K+-evoked, Ca-'+-dependent release of endogenous glutamate, and not aspartate The release of endogenous glutamate was not affected by the tetanus toxin treatment In conclusmn, this study provides clear evidence of Impaired release of the inhibitory transmitter, G A B A ,

Ex vtvo release of GABA in the active seizure phase

during a c h r o m c epileptic syndrome. Moreover, this impaired release recovered w h e n the seizures remit, although other changes in the release process do last longer and may be related to the continuing functional abnormalities in these animals Acknowledgements--We thank the British Epilepsy Research

Foundation for their support, and Dr J Mellanby for helpful discussion J G R J is a Wellcome Semor Lecturer REFERENCES Albus U and Habermann E (1983) Tetanus toxin inhibits the evoked outflow of an inhibitory (GABA) and an excitatory (D-aspartate) amino acid from parhculate brain cortex Toxicon 21, 97 1I0 Bagetta G , Knott C , Nlstlc6 G and Bowery N G. (1990) Tetanus toxin produces neuronal loss and a reduction m GABAA but not GABAB binding sites m rat hlppocampus Neurosct Lett 109, 7-12 Bergey G K , Macdonald R L , Hablg W H , Hardegree M C and Nelson P G (1983) Tetanus toxin convulsant acUon on mouse spinal cord neurons m culture J Neuro~cl 3, 2310-2323 Bevan S and Wendon L M B (1984) A study of the action of tetanus toxin at rat soleus neuromuscular juncUons J Physlol Lond 348, 1 17 Blttner M A and Holz R W (1988) Effects of tetanus toxin on catecholamme release from intact and d~gltomnpermeablhzed chromaffin cells J Neurochem 51, 451 456 Brace H M , Jefferys J G R and Mellanby J. (1985) Longterm changes m hlppocampal physiology and in learning abdlty of rats after mtrah~ppocampal tetanus toxin J Physlol Lond 368, 343-357 Colhngndge G L and Dawes J. (1982) The in vitro mhlbmon of GABA release by tetanus toxin Neuropharmacoloqy 21, 851 855 Colhngndge G L , Colhns G G S, Dawes J , James T A , Neal M J and Tongroach P (1980) Effect of tetanus toxin on transmitter release from the substantm mgra and strlatum tn vttro J Neurochem 34, 540~547 Colhngndge G L , Thompson P A , Davies J and Mellanby J (1981) In vitro effect of tetanus toxin on GABA release from rat hlppocampal shces J Neurochem 37, 1039-1041 Green A R and Vincent N D (1987) The effect of repeated electroconvulswe shock on GABA synthesis and release m regions of rat brain Br J Pharmac 92, 19-24 Green A R , Mmchln M C and Vincent N D (1987) Inhlbmon of GABA release from shces prepared from several brain regions of rats at various t~mes following a convulsion Br J. Pharmac 92, 13-18 Hablg W H , Blgalke H , Bergey G K , Neale E A , Hardegree M C and Nelson P G (1986) Tetanus toxin m dissocmted spinal cord cultures long-term char-

379

actenzatlon of form and acUon J Neurochem 47, 930~ 937 Hawkins C A and Mellanby J H. (1987) Llmblc epilepsy reduced by tetanus toxin a longitudinal electroencephalographlc study Epdepsta 28, 431-444 Jefferys J G R (1986) Tetanus toxin chromc epileptic focl m rat hlppocampal shces J Phys~ol, Lond 373, 24P. Jefferys J G R (1989) Chromc epileptic foci m vitro m hlppocampal shces from rats with the tetanus toxin epileptic syndrome J Neurophyszol 62, 458-468 Jefferys J G R and Roberts R (1987) The biology of epilepsy In Epdepsy (Hopkins A., ed ), pp 19-81 Chapman & Hall, London Jefferys J G R and Wdhams S F (1987) Physiological and behavioural consequences of seizures induced m the rat by lntrahlppocampal tetanus toxin Bram 110, 517-532 Jefferys J G R , Hughes S, Smxth B and Wdhams S F (1988) Do tetanus toxin seizures kdl hippocampal pyramidal cells9 Neuroscl Lett. Suppl 32, $66. Kamphms W , Hmsman E , Dreljer A M C , Ghljsen W E J M , Verhage M and Lopes da Sdva F H (1990) Kmdhng increases the K+-evoked Ca2+-dependent release of endogenous GABA in area CA1 of rat hlppocampus Bram Res 511, 63 70 Mellanby J H (1989) Ehminatlon of t25I from rat brain after rejection of small doses of 12Si_labelled tetanus toxin into the hlppocampus Neurosct Lett Suppl, 36, $55 Mellanby J , George G , Robinson A and Thompson P (1977) Eplleptlform syndrome m rats produced by inJecting tetanus toxin into the hlppocampus J Neurol Neurosurq Psychiat 40, 404~114 Mellanby J , Renshaw M , Cracknell H , Rands G and Thompson P (1982) Long-term impairment of learning ablhty m rats after an experimental hlppocampal epdeptlc syndrome Exp Neurol 75, 6904599 Mellanby J , Hawkins C , Mellanby H , Rawhns J N P and Impey M E (1984) Tetanus toxin as a tool for studying epdepsy J Physlol. Parts 79, 207-215 Nlcholls D G , Slhra T S and Sanchez-Pneto J (1987) Calcmm dependent and independent release of glutamate from synaptosomes monitored by continuous fluorometry J Neurochem 49, 50-57 Pellegrmo L J , Pellegrlno A S and Cushman A J (1979) A Stereotaxzc Atlas of the Rat Bram Plenum Press, New York Sanchez-Pneto J , Slhra T S and Nlcholls D G (1987) CharacterlzaUon of the exocytotlc release of glutamate from guinea pig cerebral cortical synaptosomes J Neurochem 49, 58~4 Skrede K. K and Malthe Sorenssen D (1981) Differential release of D-(3H) aspartate and (14C) gamma-aminobutyric acid following actwatlon ofcommlssural fibres in a longitudinal shce preparation of guinea pig hlppocampus Neurosct Left 21, 71-76 Szerb J C , Ross T E and Gurewch L. (1981) Compartments of labeled and endogenous ),-ammobutync acid giving rise to release evoked by potassium or veratrldme m rat cortical slices J Neurochem 37, 1186-1192