Low extracellular magnesium unmasksN-methyl-d -aspartate-mediated graft-host connections in rat neocortex slice preparation

Neuroscience Vol. 64, No. 2, pp. 443 458, 1995

~)

Pergamon

0306-4522(94)00372-6

Elsevier Science Ltd Copyright F 1994 IBRO Printed in Great Britain. All rights reserved 0306-4522/95 $9.50 + 0.00

LOW E X T R A C E L L U L A R M A G N E S I U M U N M A S K S N-METHYL-D-ASPARTATE-MEDIATED GRAFT-HOST C O N N E C T I O N S IN RAT N E O C O R T E X SLICE PREPARATION V. V. SENATOROV,*t:~ I. VILAGI,§ I. T A R N A W A , § I. B A N C Z E R O W S K I - P E L Y H E § and Z. FULOP~[I fPavlov Department of Physiology, Institute for Experimental Medicine, St. Petersburg, Russia §Department of Comparative Physiology, E6tv6s Ldr;ind University and :~First Department of Anatomy, Semmelweis University Medical School, Budapest, Hungary

Abstract--The main purpose of this study was to investigate the role of N-methyl-D-aspartate receptors in host graft synaptic transmission in the neocortex. The effects of low extracellular magnesium, the glutamate agonist N-methyl-D-aspartate and N-methyl-D-aspartate antagonists on the synaptic activation of connections between embryonic neocortical graft tissue and the surrounding host tissue were studied in 17 perfused slices of rat neocortex. In standard artificial cerebrospinal fluid, stimulation of the host white matter evoked field potentials in four of 17 grafts. However, in Mg2+-free medium, the same stimulation evoked field potentials in an additional six grafts, with significant increases in the mean duration of the evoked responses in the 10 responsive grafts. In five of these slices stimulation of the graft also evoked field potentials in the host tissue, suggesting reciprocal interaction between graft and host. Simultaneous extracellular recordings from graft and host tissues in Mg2+-free medium showed that spontaneous epileptiform discharges developed in the graft and host tissue synchronously. In Mg2+-free medium, application of N-methyl-D-aspartate induced a shift of the baseline with superimposed epileptiform discharges in both graft and host. Application of the non-competitive N-methyl-D-aspartate antagonist ketamine and the competitive antagonist D,L-2-amino-5-phosphonovaleric acid attenuated or reversibly blocked both the spontaneous epileptiform discharges and the evoked field potentials. Our data provides evidence that N-methyl-o-aspartate receptors are present at synapses created between fetal graft and host neocortex, and that the N-methyl-D-aspartate-activated receptor~zhannel complex plays an active role in mediating excitatory synaptic transmission in host-graft circuitry.

A growing body of evidence indicates that homotopically transplanted fetal neocortical tissues can be both functionally and structurally integrated into the host neocortex to form viable connections with the h o s t brain. 3"5"8"13"17'28'41"43"44"49 53 While much of our

Electrophysiological experiments in anaesthetized animals have demonstrated that fetal neocortical tissue transplanted into different neocortical areas receives functionally active inputs from the host brain. 5'17"28'4I'51"6° However, these in vivo studies proknowledge about the connections between the vided little information about efferent connections neocortical transplant and the host neocortex tissues and the transmitters that mediate host-graft comhas been obtained from anatomical studies, 8"13'14'43'44,49 munication. The complexity of the experimental little is known about the mechanisms by which grafts conditions (e.g. anaesthesia, stress) and difficulty in functionally integrate into the host neocortex. positioning the recording electrodes precisely have also hindered interpretation of the in vivo data. ~8 *To whom correspondence should be addressed. Present These difficulties can be overcome by using an in vitro address: Neuroscience Department, Loeb Research Inbrain slice preparation containing both the transplant stitute, Ottawa Civic Hospital, 1053 Carling Avenue, and the surrounding host tissue. 16,18,21,40,48Such studies Ottawa, Ontario K1Y 4E9, Canada. ]lPresent address: Brain Research Laboratory, Institute of have demonstrated that transplanted immature neurAnimal Behaviour, Rutgers, The State University of ons possess electroresponsive properties similar to New Jersey, 10l Warrent Street, Newark, NJ 07102, those seen in the same type of neurons in a normal U.S.A. brain. In addition, stimulation of the host tissue Abbreviations: ACSF, artificial cerebrospinal fluid; APV, l~,L,-2-amino-5-phosphonovaleric acid; FP, field poten- resulted in a response in the transplant and vice versa, tial; HEPES, N-2-hydroxyethylpiperazine-N'-2-ethone suggesting the presence of functioning reciprocal sulphonic acid; MK-801, {(+)-5-methyl-10,11-dihydroconnections between the two a r e a s . 16,~8-2~'4°'48 5H-dibenzo[a,d]cyclohapten 5,10-imine moleate}; In spite of a great deal of morphological and NMDA, N-methyl-D-aspartate; PB, phosphate buffer; PBT, phosphate buffer with Triton X-100; SED, electrophysiological evidence for the formation of a spontaneous epileptiform discharge. functional host-graft connection in the neocortex, 443

V.V. Senatorov et al.

444

few studies have dealt specifically with the p h a r m a cology of neocortical graft connections. 28'5~ Excitatory a m i n o acids, primarily L-glutamate a n d L-aspartate, are the m a j o r excitatory n e u r o t r a n s m i t ters in the m a m m a l i a n telencephalon (for review see Ref. 61). Excitatory synaptic transmission activated by g l u t a m a t e is m e d i a t e d by b o t h N-methyl-D-aspartate ( N M D A ) a n d n o n - N M D A receptors. 6'24'32 The n o n - N M D A receptor--channel complex mediates fast v o l t a g e - i n d e p e n d e n t synaptic conductance, while N M D A receptors play a role in the generation o f slow voltage-dependent excitatory postsynaptic potentials. 6,24,29'32'35'42 Activation of the N M D A r e c e p t o r - c h a n n e l complex is governed by the presence o f m a g n e s i u m ion (Mg2+). 2"6"24'29'32"34"35'37"42"55 To c o n d u c t ions efficiently when activated by glutamate, the N M D A c h a n n e l m u s t be depolarized to a voltage range in which M g 2+ is removed from the channel. C o n s i d e r i n g the wide distribution o f N M D A receptors in neocortical tissue (for review see Ref. 1 1) a n d their i n v o l v e m e n t in n e u r o n a l d e v e l o p m e n t a n d synaptic plasticity, 7'3~'46 we postulated that the N M D A ion c h a n n e l - r e c e p t o r complex m i g h t play an i m p o r t a n t role in mediating functional connections of the intracortical transplant. In the present study we investigated the effects o f low extracellular m a g n e s i u m on the functional expression o f the N M D A receptor in h o s t - g r a f t c o n n e c t i o n s in a b r a i n slice preparation. The slice c o n t a i n e d b o t h the t r a n s p l a n t a n d the s u r r o u n d i n g host tissues. Evoked field potentials (FPs) and spontaneous epileptiform discharges (SEDs) were recorded simultaneously from host and graft p o r t i o n s o f the slice.

dissected and 400/~m coronal slices were cut with a Vibratome (Fig. l). Usually three or four neocortical slices containing the graft were prepared and the one that contained the largest graft area was chosen for electrophysiological investigation (Figs 1, 2). After a 30min preincubation in oxygenated HEPES-buffered solution at room temperature, a slice was transferred to an interfacetype slice chamber. It was perfused at a rate of 3 ml/min with either standard artificial cerebrospinal fluid (ACSF) containing (in raM): 126NAC1, 1.8KC1, 1.25KH2PO 4, 1.3 M g S O 4, 26NaHCO 3, 2.4CAC12, 10 glucose, or with Mg2+-free ACSF (as above except that magnesium was omitted). The temperature was regulated at 34 + 0.YC. All drugs were delivered to the brain slices through the perfusion line.

Histology To illustrate the structure and orientation of neural fibres in the grafted tissue in our slice preparations, we used neurofilament immunohistochemistry. At the end of each recording session, the slices were fixed with 2% paraformaldehyde in 0.1 M phosphate buffer (PB; pH 7.4) for two days and photographs were taken under a dissecting microscope to record their gross anatomical features. The slices were then further prepared for specific neurofilament staining. Vibratome sections (20/~m) were picked up in PB and treated with 20% normal goat serum for 1 h at room temperature. Free-floating frontal sections were then washed in PB and incubated with monoclonal antibody against the 160,000mol. wt neurofilament polypeptide (I : 100 dilution) in PB containing 0.5% Triton X-100 (PBT) for 48h at 4°C. After a washing in PB, sections were incubated with anti-mouse biotinylated antibody (1:100 dilution) in PBT for 2 h at room temperature, washed in PB and then incubated with streptavidin-horseradish peroxidase conjugate (1:100 dilution) in PBT for 2 h at room temperature. Following washing in 0.05 Tris buffer (pH 8), the immunocomplex was visualized by reacting with 0.03% diaminobenzidine and 0.3% NiNH4SO4 in Tris buffer for 30 min in the dark at room temperature. Concentrated H202 (15/~1/1 ml solution) was then added and incubation was continued for an additional 4 min. After washing, sections were mounted on gelatine-coated slides, dehydrated with graded ethanols, cleared with xylene and coverslipped with mounting media for examination under the microscope.

EXPERIMENTAL PROCEDURES

Transplantation Adult female Carworth Farm Y rats (Laboratory Animal Institute, G6d6116, Hungary) weighing 200-250 g were used as graft recipients. Under general anaesthesia (a ketamine hydrochloride and xylazine mixture), a 5 × 5 mm burr-hole was made in the skull in such a way that the lateral edge of the bone patch remained attached to the skull, leaving the periosteum undamaged. A small (2.0 × 1.5 mm) fragment of the left somatosensory cortex just rostral to bregma was then removed by aspiration. The use of a door-like opening in the skull maintained some blood supply to the bone and allowed subsequent covering of the underlying brain after the surgery. After a five to seven day recovery period, a small block (1 × 1 mm) of tissue obtained from the middle portion of the rostral neocortex of a fetal rat (embryonic day 15) was transplanted into the lesion cavity of the adult recipient. Slice preparation Two to three months after the transplantation, the host rats were decapitated under ether anaesthesia. The brains were removed and placed in ice-cold, oxygenated (95% 02 , 5% CO2) HEPES-buffered solution (composition in mM: 120 NaCI, 2 KCI, 1.25 KH2PO 4, 2 MgSO 4, 20 NaHCO3, 2 CaC12, l0 glucose, 6.7 HEPES-acid, 3.3 Na-HEPES). A block of the sensorimotor cortex containing the graft was

Recording and stimulation Two broken-tipped micropipettes, filled with l M NaCI (d.c. impedance of 10 15 Mf~) were positioned to record concurrently from: (i) the transplant; and (ii) in the host cortical tissue (layers II-III). A glass-insulated bipolar platinum stimulation electrode was placed; (i) at the border between the white and the gray matter of the host (in all of 17 preparations); or (ii) in the deepest part of the graft (in five of 17 preparations) (Fig. 1). Stimuli consisted of single shocks of 0.2ms given at 0.1, 0.2, 0.5 and 1 Hz. Stimulus intensity (2-7 V) was about 1.5 times threshold for evoking a field potential. Optimal placement of the recording electrodes in both host and graft was determined by varying their positions so that the largest field response was recorded following stimulation of the host or graft regions. Signals were conventionally filtered, amplified and monitored on a digital oscilloscope, with concurrent strip-chart recording. Latency, amplitude and duration measurements were made with the oscilloscope. The FP amplitude was taken as the voltage difference between the peak positive deflection and the negative deflection. The amplitudes of the positive and negative components were taken as the maximal deflection from baseline. All data are expressed as mean _+ S.E.M. and Student's t-test was used for data analysis. Synchronization of seizure activity between host and graft tissues was defined as a I00% simultaneity of spontaneous FPs in host and graft during a 30 min period.

NMDA receptor-mediated graft-host connections in the neocortex

Experimental protocol

445

RESULTS

Seventeen successful experiments were carried out on cortical slices taken from the brains of 17 host animals. All slices contained both the graft and the host tissue, and concurrent paired recordings from both graft and host parts of the slice were made in each experiment. Experimental procedures were carried out in the following steps. (1) Slices were perfused by standard ACSF. Stimulation of the host white matter was used to evoke FPs from both the host and graft portions of the slice. (2) The perfusate was changed to Mf+-free ACSF. Stimulation of the host white matter was used to evoke FPs from both the host and graft portions of the slice. (3) The stimulating electrode was moved from the host tissue to the graft and standard ACSF was reintroduced. Stimulation of the graft was then used to evoke FPs from both the host and graft portions of the slice. (4) Standard ACSF was switched to Mg2+-free solution. Evoked FPs from both the host and graft portions of the slice were recorded following stimulation of the graft. Steps 1 and 2 were performed in all 17 experiments; steps 3 and 4 were done in five of the 17 experiments.

Materials NMDA, D,e-2-amino-5-phosphonovaleric acid (APV) and ketamine were purchased from Sigma Chemical Co., U.S.A. Monoclonal antibody against 160,000 tool. wt neurofilament polypeptide, anti-mouse biotinylated antibody and streptavidin horseradish peroxidase conjugate were purchased from Boehringer Mannheim Biochemica, Vector Laboratories Inc., U.S.A. and Amersham International, respectively,

Anatomical observations To o b t a i n morphological correlates of our electrophysiological findings we performed histological examinations o f all slice p r e p a r a t i o n s studied. All the host neocortical tissues examined h a d well-developed grafts that always filled the available space in the lesion cavity. The unstained thick sections prepared for the electrophysiological study showed welldefined patterns of gray a n d white matter. Each graft had its own internal fibre network seen as a white tangled substance connecting different parts of the transplant. However, the structure and the orientation of the graft tissue were different from those of the host cortex (Fig. 2). The unstained light microscopic picture showed a variation in the organization of the network of fibre fascicles in the graft portions of different slices. The structure consisted of fibre fascicles r u n n i n g in one direction (Fig. 2A, B), or spherically r u n n i n g surface bundles connected by radially oriented internal fascicles (Fig. 2C). M o r p h o logical analysis of sections of these slices stained for neurofilament protein provided evidence that the white bundles seen in the slices were indeed neural (Fig. 3).

7 1

Graft stimulation 5/17 slices

t 2

\ f

Host stimulation

17slices

Fig. 1. Experimental paradigm. (1) Dissection of a thick (4-5 mm) slice of the host brain containing the graft. (2) Trimming the block of somatosensory neocortex to isolate the graft and surrounding host tissue. (3) Slicing to a thickness of 400 ~m. (4) Stimulation of the host and concurrent recording from the host and the graft in standard and Mg2+-free ACSF. (5) Stimulation of the graft and concurrent recording from the host and graft in standard and Mg 2÷-free ACSF.

Fig. 2. (Caption opposite).

NMDA receptor-mediated graft-host connections in the neocortex

447

Fig. 3. A photomicrographic montage showing an anti-neurofilament immunostained section. Note bundles of neural fibres connecting different parts of the graft. Arrowheads, graft host border; arrows, bundles of neural fibres; T, transplant; NC, neocortex of the host; cc, corpus callosum. Scale bar = 150 # m.

To characterize structural integration of the graft tissue with the host neocortex in our preparations, the region of the graft-host border was examined to compare morphological and electrophysiological findings. A sharp, clearly-defne d boundary (Fig. 4A) could only be observed in sections prepared from the seven slices where no functional connections were found in electrophysiological experiments. On the contrary, no clearly outlined border was observed in histological preparations from the 10 slices with functional host-graft connections. In sections pre-

pared from these slices numerous neurofilamentpositive fibres crossing the host-graft border were seen under higher magnification, showing reciprocal innervation of the host-graft tissue (Fig. 4B, C).

Stimulation of the host white matter Stimulation of the host tissue at the border of the white and the gray matter in the 17 slices (Fig. 1) initiated FPs that were recorded in: (i) layers I I - I I I in the host neocortex; and (ii) at the same level in the graft. The shape of the potentials was variable, but

Fig. 2. Representative photographs of slices containing the graft. Note the presence of a well-developed network of neural fibres (arrows) seen as a white tangled substance running in different directions all over the transplant. Arrowheads, graft-host border; T, transplant; nc, neocortex of the host; cc, corpus callosum.

448

V . V . Senatorov et al.

Fig. 4. Anti-neurofilament immunostaining showing neural fibres crossing the h o s t ~ r a l t border to create a common neuropil. Arrowheads, h o s t ~ r a f t border; arrows, neural fibres crossing the h o s t ~ r a f t border: T, transplant; nc, neocortex of the host. Scale bar = 25 #m.

NMDA receptor-mediated graft host connections in the neocortex

449

Control V

J0,mv 200 r't~ec

HOST

Mg "~ .

GR,ab'I

HOST

Fig. 5. High speed chart recordings of evoked FPs simultaneously recorded in host and graft following electrical stimulation of the host (arrowhead) in standard and MgZ+-free ACSF. Two successive components could be distinguished: a shorter early-positive component and a longer late-negative component. Switching to Mg2+-free ACSF significantly increased the negative component of the evoked FP both in the host and in the graft. Note the epileptiform or rhythmic discharges seen in the late (negative) component.

two successive c o m p o n e n t s could be distinguished in most FPs: an early-positive a n d a late-negative component. The positive c o m p o n e n t h a d a lower amplitude and shorter d u r a t i o n t h a n did the negative Control

60min

//

vo.~ v H, T vO-2H~oHzY o.; V Hz V V vvv

c o m p o n e n t (Fig. 5). The threshold stimulus was not different in the host a n d graft portions of the slice (3.46 _ 0.30 and 3.05 _+ 0.41 V, respectively) but the onset latency of the FPs in the host (6.25 _+ 0.45 ms)

Mg++.free 0.1Hz v ~OHz V0.1Hz V r V0.2Hzv NMDA ~

r ..............

]

!

HOST

I

]

'

0.5s

I

]

1

0.5 mV

5rain

Fig. 6. Chart recording of extracellular potentials recorded concurrently from the graft and host portions of the neocortical slice. Stimulation of the host (double arrowhead) with frequency 0.1-1.0 Hz elicited evoked FPs simultaneously in both graft and host in standard ACSF (control) and later in MgZ+-free ACSF. Note that after switching to Mg2+-free ACSF and in the absence of stimulation, spontaneous epileptic FPs appeared and discharged synchronously in the host and graft. NMDA application (10/~ M) enhanced the frequency of spontaneous FPs in both graft and host parts of the slice and this response mimicked evoked FPs elicited by electrical stimulation of the host. One pair of spontaneous epileptiform FPs is shown at a faster time-base in the inset. NSC

64,2 (;

/

450

V. V. Senatorov 60min

Control

//

et al.

M g ÷÷ - f r e e 0.1Hz 0.1 Hz YYY ¥0.2Hz VVyVVVVV ¥ V ¥ u. ..o n. Z. ~ .u ~ u,SHz¥0.2Hzvv ¥ ¥

0.1 HZ VVVVVVVV

,

, v,

~,~¢,

¥,

NMDA

Y

r

,

l

I

I

1

1

'

'

!

GRAFT

I

m

L__ J '""I HOST

0.5 mV 5 rain

Fig. 7. Chart recording of extracellular potentials recorded concurrently from the graft and host portions of the neocortical slice first in standard and then in MgZ+-free ACSF. Note that in standard ACSF, stimulation of the host (double arrowhead) with frequency 0.1-1.0 Hz elicited evoked FPs in the host but not in the graft. On the other hand, after switching to the Mg 2+-free ACSF, stimulation of the host elicited evoked FPs not only in the host but also in the graft. Note that the graft cannot follow host stimulation to as high a frequency as can host tissue, the difference appearing at 0.5 and 1 Hz. One pair of evoked FPs is shown in the inset with an extended time-base. was significantly shorter than that in the graft (14.43 + 0.53 ms, P < 0.001). In control (standard ACSF), evoked FPs could be recorded simultaneously in host and graft tissues in only four out of 17 slices (Fig. 6). In the other 13 slices, stimulation of the host white-gray border evoked FPs from the host part of the slice, but not from the graft. Stimulation of the white-gray matter boundary of the host tissue was more effective in evoking FPs in the graft in Mg2+-free A C S F . One hour after switching from standard A C S F to Mg2+-free medium, evoked FPs were recorded simultaneously from both the host and the graft tissues in 10 of 17 slices (these 10 include the four slices in which the transplant responded in standard A C S F ) (Figs 6, 7). Changing to Mg2+-free A C S F had no effect on the stimulus threshold for activation of FPs either in the host or in the graft parts of the slice (3.36_+ 0.27 V and

3.48 _+ 0.43 V, respectively). The onset latency of the FPs did not change significantly in Mg2+-free A C S F and continued to be twice as long in the graft (15.33 + 0.33 ms) as it was in the host (6.20 _+ 0.74 ms, P < 0.001). Observable changes of amplitude and duration of FPs in MgZ+-free A C S F occurred mainly in the late (negative) component of the FP (Fig. 5, Table 1). Mg2+-free A C S F did not significantly change the amplitude of FPs in the graft, but in the host the amplitude of FPs was almost twice that in standard A C S F (Table 1). In both host and graft the duration of the FPs was about eight times greater than in standard A C S F (Table 1). In addition to the increase in duration and amplitude, the late component also became epileptiform (composed of several spike- or wave-like discharges rising from a long-lasting negative deflection; Fig. 5):

Table 1. Effect of host stimulation: Mg2+-free artificial cerebrospinal fluid-induced changes in amplitude and duration in the two components (early-positive and late-negative) of field potentials evoked by white matter stimulation Standard ACSF

Mg2+-free ACSF

Host recording Graft recording Host recording Graft recording (n = 17/17) n = 4/17) (n = 17/17) (n = 10/17) Amplitude (mV)

Early (positive) component Late (negative) component

0.18 _+0.04 0.45 + 0.07

0.19 _+0.09 0.48 _+ 0.11

0.23 _+ 0.04 0.90 _+_0.16*

0.20 + 0.05 0.74 + 0.14

Duration (ms)

Early (positive) component Late (negative) component

68 + 23 263 _+ 34

41 + 12 217 + 65

199 + 85 1632 + 180"*

137 _%32* 1569 + 199"*

Results are expressed as mean + S.E.M. Asterisks indicate that the parameters measured in Mg2+-free ACSF are significantly different from the same parameters measured in standard ACSF: *P < 0.05; **P < 0.001. Note that the only significant amplitude change was that of the late (negative) component of the host, which increased in zero Mg 2+, but the duration of the late component was significantly increased at both sites in zero Mg 2+.

NMDA receptor-mediated graft host connections in the neocortex These evoked FPs could be recorded from the entire area of the graft, suggesting synchronous activation of the graft neurons. In all but two experiments, FPs evoked in the graft by host stimulation could not follow repetitive stimulation at a frequency as high as 1 Hz (Fig. 7). In seven out of 17 slices, electrical stimulation of the host evoked FPs only in the host and not in the graft, regardless of whether standard or Mg2+-free A C S F was present (data not shown).

Stimulation of the graft To detect the presence of reciprocal connections between graft and host, we conducted tests with stimulation of the graft. For this purpose, we chose five slices in which previous stimulation of the host tissue had evoked FPs in the host and graft in either standard or Mg2+-free ACSF. The stimulation electrode was positioned in the graft and the FPs elicited by stimulation of the graft were recorded concurrently in both graft and host, as described above (Fig. 1). During perfusion with standard A C S F , stimulation of the graft elicited graft and host FPs simultaneously in four out of five slices (Fig. 8). In the remaining

,v

Host

-

0.1HZ t t 1,0.2

slice, evoked FPs were recorded only in the graft. The appearance of the FPs elicited by graft stimulation was similar to those recorded following stimulation of the host and consisted of the same two major components: an early-positive and a late-negative component. The stimulus threshold for activation was not different in the host and the graft portions of the slice (2.50 _+ 0.54 and 3.20 _+ 0.83 V, respectively) but the onset latency of the FPs in the host (14.50 +_ 2.10 ms) was approximately two-fold greater than in the graft (6.50 __ 1.19 ms, P < 0.05). After switching the standard perfusate to an Mg 2+free one, FPs could be recorded from the host in one more slice, i.e. stimulation of the graft initiated simultaneous FPs from the host and the graft tissues in all five slices studied. Switching to Mg2+-free A C S F did not significantly change the onset latency or stimulus threshold in either the host or graft ( 1 3 . 5 0 + 2 . 1 0 m s , 2.00_+0.61V and 6.50_+ 1.19ms, 1.80_+ 0.57V, respectively). However, there was a doubling of the amplitude of the negative component of the FPs in the host tissue, and the duration of this component in the host and the graft increased by nine and five times, respectively (Table 2). In all five slices, FPs recorded in the host tissue did not follow the

stimulation

Control

60rain

//

~

Mg++.free

60rain

//

0.1Hz

¥10.5

¥¥ I.OY¥

¥ t ~¥¥

,,,

,%s, !.o ,,,

,' '

451

Wash

-

Graft

stimulation

Mg++-free

/

0.1Hz02

.2

11 lll°i s,.o.,| 111|

,'i °, I

HOST

Y U 0.Ss

0.5 mV

5 min

_ ,/f J/

t

r Fig. 8. Chart recording of extracellular potentials recorded concurrently from the graft and the host portions of the neocortical slice. Stimulation of the host (double arrowhead) elicited evoked FPs simultaneously in the graft and host in standard (control) and later in Mg2+-free ASCF. Switching the stimulation to the graft (double arrow) also elicited evoked FPs simultaneously in the graft and host in standard (control) and later in Mg2+-free ACSF. Under each pair of sweeps are examples of a pair of evoked FPs shown with an extended time-base.

V.V. Senatorov et al.

452

Table 2. Effect of graft stimulation: Mg2+-free artificial cerebrospinal fluid-induced changes in amplitude and duration of the two components (early/positive and late/negative) of evoked field potentials Mg2+-free ACSF

Standard ACSF

Host recording Graft recording Host recording Graft recording (n = 4/5) (n = 5/5) (n = 4/5) (n = 5/5) Amplitude (mV)

Early (positive) component Late (negative) component

0.23 _+0.07 0.21 ± 0.02

0.17 _ 0.02 0.64 ± 0.16

0.30 ± 0.07 0.74 ± 0.21"

0.20 __+0,08 0.78 ± 0.21

Duration (ms)

Early (positive) component Late (negative) component

38 ± 11 115 ± 16

25 ± 8 223 _+ 88

104 _+26 1406 ± 397*

128 ± 31" 1360 _+ 73**

Results are expressed as mean + S.E.M. Asterisks illustrate the significant changes in Mg2+-free ACSF: *P < 0.05; **P < 0.001. Note that the amplitude of the late (negative) component of the host was significantly increased in zero Mg 2÷, but the amplitude in the graft was not. The durations of both early and late components were significantly prolonged at both sites in zero Mg 2+.

repetitive stimulation of the graft at frequencies as high as 1 Hz (Fig. 8).

Spontaneous epileptiform discharges in Mg2+-free artificial cerebrospinal fluid Ten to twenty minutes after changing the standard A C S F to an Mg2+-free medium, spontaneous field potentials, i.e. SEDs, developed in all slices (Figs 6, 7). The amplitude of the SEDs ranged from 0.5 to 4 mV and their duration varied from 100 to 1200 ms. The SEDs usually developed first in the host part of the slice and only later in the graft. The SEDs consisted of two major deflections, an early-positive and a late-negative, and their duration and amplitude varied considerably. In some recordings the

NMDA

positive component was poorly expressed or absent (Figs 6, 9). Superimposed population spikes were often present in the second (negative) component (inset in Fig. 6). In spite of some variability in the form, amplitude and duration of the SEDs, the appearance of epileptiform discharges in graft and host tissues was similar. In the 10 experiments where stimulation of the host white matter in Mg2+-free solution resulted in an evoked potential in both parts of the slice, epileptic discharges appeared with a mean frequency of 1.61 + 0.23 Hz. Each epileptic discharge in the host was paired with a discharge in the graft (Fig. 9). In Mg2+-free A C S F , SEDs could either develop synchronously in the two parts of the slice from the very NMDA

GRAFT

0.5 mV 5 mln

Fig. 9. Chart recording of extracellular potentials recorded concurrently from the graft and host portions of the neocortical slice in Mg2+-free ACSF. Note the synchronicity between the SEDs recorded in the host and graft. Bath application of NMDA (10 and 15/~M) raised SED frequency in both parts of the slice.

453

NMDA receptor-mediated graft-host connections in the neocortex beginning, or start independently in the graft and host and become synchronous 5-10min later, after the SEDs had gained a certain stable level of magnitude and regularity. In seven out of 17 slices where no host-graft functional connections were found by electrical stimulation, SEDs appeared independently in the host and graft parts of the slices, and with a discharge frequency in the host tissue up to three times (1.87_+ 0.25 Hz) that in the graft (0.55 +_ 0.07 Hz, P < 0.001) (data not shown). The SEDs could be recorded from the whole area of the graft. In some slices, large (5-13 mV) spontaneous spreading depression-like negative potentials developed during incubation in the Mg > free medium (data not shown). Unlike the SEDs, spontaneous spreading depression appeared independently in host and graft tissue.

Application antagonists

of

N-methyl-D-aspartate

and

its

To detect the presence of NMDA receptors in the neurons of the graft, we studied the sensitivity of the graft tissue to NMDA. For this purpose, we applied NMDA in all 17 experiments. During perfusion with Mg2+-free ACSF, bath application of NMDA

(5-15/~M) induced a significant increase in the frequency of SEDs, which was usually coupled with a d.c. shift of baseline (Figs 6, 9). We found that the responses to NMDA application developed in the graft and the host portions in all slices tested, independently of the presence of functional connections between them. The NMDA-initiated increase in SEDs was dose-dependent (Fig. 9). To study a possible involvement of NMDA receptor-channel complex activation in the mediation of host-graft connectivity, we investigated the impact of NMDA receptor antagonists on synaptic transmission between the two tissues. Two selective NMDA antagonists were chosen for application in Mg2+-free ACSF in slices where recording of evoked FPs revealed functional host-graft connections: the non-competitive NMDA antagonist ketamine and the competitive antagonist APV. Application of ketamine (20 # M, n = 5) influenced both the SEDs and the FPs evoked by electrical stimulation of the host white matter. Ketamine application reversibly blocked the SEDs and abolished their synchronous rhythm in graft and host tissues for up to 20 min (Fig. 10). It also blocked FPs evoked in the graft by host white matter stimulation and significantly decreased the amplitude of evoked FPs in the host

Ketamine O.IHz

¥¥¥¥¥~

0.1Hz

0.1Hz

VVVV¥¥

tV~VYV

HOST 0.5 mV

5 min w

?

• il q

~ii '

GRAFT

Fig. 10. Chart recording of extracellular potentials recorded concurrently from the graft and host portions of the neocortical slice in Mg2+-free ACSF. Application of ketamine (20/~M) blocked spontaneous epileptiform FPs which discharged synchronously in host and graft parts of the neocortical slice. Also blocked were'FPs evoked in the graft by host stimulation (double arowhead). FPs evoked in the host were significantly decreased in amplitude.

V.V. Senatorov et al.

454

Fig. 10). Addition of APV (20#M, n = 6 ) had a similar effect. It not only reversibly blocked SEDs developing in Mg2+-free ACSF but also prevented N M D A application from causing a burst of epileptiform discharges (Fig, 11). This effect lasted as long as 30 min. APV application significantly decreased the amplitude of the FPs evoked in both host and graft tissues when stimulating the host (Fig. 11). These experiments showed that specific NMDA receptor antagonists selectively blocked synaptic transmission between host and graft portions of the neocortical slice. DISCUSSION

Structural pathways within the graft and at the border with the host In the present study we used unstained preparations and neurofilament immunostaining to demonstrate the presence of a well-developed fibre network within the graft. The presence of a special neuronal network responsible for the synchronization of the activity of neurons may contribute to the development and spreading of epileptic processes in the intact neocortex. Recurrent excitatory synaptic circuits are thought to be the primary synchronizing mechanism in the spread of epileptiform activity.9'1°'55'59 Strong connections within neocortical transplants have also been shown using horseradish peroxidase 14"44 and Phaseolus vulgaris leucoagglutinin. 49 Such networks could be the anatomical substrate of the synchronous bursting of a large population of neurons inside the graft that is necessary for the generation of both evoked FPs and SEDs.

The presence of structural pathways interconnecting graft and host tissues, as well as neurons within the graft, is now well documented. 8'43'44'49 Our immunohistochemical results show that a great number of neural fibres crossed the graft host border, but only in the slices where functional connections between two tissues were found. This finding corroborates our physiological experiments.

Reciprocal host-graft connectivity determined by recording evoked field potentials Electrophysiological responses in the graft following host stimulation, and in the host following graft stimulation, suggested that functional reciprocal connections had been established between the graft and the host neocortical tissues. The presence of reciprocal graft connections has previously been demonstrated in slice preparations obtained from other brain regions, e.g, lateral-geniculate nucleus transplanted into visual neocortex, septum grafted into a transected fimbria-fornix and hippocampal neurons transplanted into hippocampus or cerebellum. 18"21'48 The relatively long latency of evoked FPs in our experiments suggests that these responses reflect multi-synaptic orthodromic activation of the graft and host neurons. The marked difference in the latencies of FPs recorded in the stimulated portion of the slice (whether host or graft) and in adjacent tissue suggests synaptic delays in signal transmission between host and graft, possibly at the border. As our signals were low-pass filtered, a considerable part of the small, early, non-synaptically generated components, representing fibre excitation, ~'2v'54 was filtered out. Based on previous analyses of the APV

NMDA

NMDA 0,1 HZ

1¥V

0.1Hz IV

0.1Hz

¥y

O,T Hz

YYYVYY

GRAFT

HOST

l!

I

0.5 mV

5 min

Fig. 11. Chart recording of simultaneous extracellular potentials recorded in Mg2+-free ACSF in the host and graft parts of the same slice. NMDA application (10/~M) enhanced the frequency of SEDs while preserving host-graft synchronicity, and this response mimicked the train of FPs evoked by electrical stimulation of the host (double arrowhead). Note that APV application (20 #M) inhibited FPs evoked by host stimulation, blocked spontaneous epileptiform FPs and blocked the effect of NMDA.

NMDA receptor-mediated graft~ost connections in the neocortex origin of the various components of evoked FPs in the neocortex, ~'27'54we suggest that both the positive and the negative components recorded in our experiments reflect excitatory responses of neurons located at various depths in the neocortex. The earlier (positive) component, which was variable in appearance, represents the first wave of cortical excitation following the current pulse. 54 It could be generated partly through synaptic mechanisms and partly through antidromic activation.27'54 The late (negative) component recorded in all FPs is caused by excitatory synaptic current sinks and may therefore depend only on synaptic mechanisms. 1-27

Activation o f host-graft connections in low extracellular magnesium To demonstrate the presence of the NMDA receptor-channel complex in the graft, we applied NMDA in Mg2~-free ASCF. We found responses to NMDA application in all transplants studied, which proves that accessible NMDA receptors are abundant in graft neurons. The NMDA receptor is a complex protein containing at least four modulatory binding domains. These include a co-activator site, which binds glycine, ~523'3~ a site within the channel that binds the open-channel blockers ketamine and MK801,22'253°'33an inhibitory Zn 2+ binding site 15'45and an Mg 2+ binding site. 34"35"37"42 In the present study we observed that withdrawing extracellular Mg 2÷ from the perfusate increased the expression of host-graft neuronal transmission. After switching from standard to Mg2+-free ACSF, electrical stimulation uncovered functional host-graft connections in six more slices. Functional connections of these six grafts could therefore be expressed only when excitatory mechanisms were potentiated by the removal of Mg 2+ ions from the external medium. The lack of a significant change in the stimulus threshold, as well as a remarkable difference between the onset latencies recorded in host and graft tissues, suggest that these additional FPs were not due to a lowered threshold in Mg2+-free ACSF having permitted triggering by direct electrical current stimulation. Since extracellular Mg 2+ can effectively block the actions of NMDA through a voltage-dependent mechanism without any significant effects on responses evoked by non-NMDA agonists, we suggest a specific role for the NMDA receptor~hannel complex in the observed activation of host-graft connectivity following the switch to a magnesium-free perfusate. In Mg 2+free solutions, NMDA agonists open cationic channels that conduct both Na + and Ca '+, while the addition of Mg 2+ produces a dramatic reduction of the NMDA-induced current and a nearly total block at the resting membrane potential at physiological concentrations (1-2 raM). z'19'34Specific non-competitive (ketamine) 19'26'3°,57 and competitive (APV) 19,57,5s NMDA receptor antagonists reversibly blocked FPs in host tissue as well as FPs recorded simultaneously in the graft during electrical stimulation of the host,

455

thereby confirming a direct involvement of NMDA receptors in mediating host-graft glutamatergic synaptic transmission. We suggest that the additional evoked FPs seen upon perfusion with Mg2+-free medium were due to an NMDA-activated potentiation of synaptic transmission between host and graft. Studies of chemical mediators that contribute to evoked FPs recorded in neocortical slices indicate that non-NMDA and NMDA receptor activation are the main contributors to the source/sink current profile underlying the biphasic appearance of the FPs. 4 The five to nine-fold increase in duration of the late negative component of the FP observed in both host and graft tissues in Mg2+-free ACSF could be caused by the NMDA component of the response. One of the important functional features of glutamate-mediated synaptic responses in the CNS is the remarkably long duration of the NMDA component, which in the absence of external Mg 2÷ may last for hundreds of milliseconds, due largely to its slow channel kinetics.6'29 In addition, the long duration and the jagged pattern of the late components of FPs observed in Mg2+-free ACSF may be due to recurrent volleys of synaptic activation. The frequency sensitivity of FPs recorded in Mg2+-free ACSF (failure to follow repetitive stimulation to frequencies as high as 1 Hz) could be due to the involvement of multisynaptic pathways between host and graft.

Synchronization o f spontaneous activity between host and graft parts o f the slice in low extracellular magnesium We rarely observed any spontaneous electrical activity in neocortical slices perfused with standard ACSF, but in Mg2+-free media, spontaneous seizure activity developed in all cortical slices, as reported previously. 1z39"47'56 Incubation in Mg2+-free ACSF induced SEDs in both host and graft parts of all slices. In seven out of 17 slices where no host-graft functional connections were found, SEDs were generated independently in the host and graft parts of the slice. In contrast, 10 slices that had functional host-graft connectivity as determined from evoked FP activity exhibited SED synchrony between host and graft, suggesting the propagation of epileptiform activity between the two tissues. The fact that in some slices synchronization of SEDs in host and graft tissues appeared only after a latent period (during which period the magnitude and regularity of SEDs increased) suggests that some form of "kindling''38 may be required for the propagation of epileptiform discharges between the two tissues. FPs reflect the response of a synchronously activated neuronal population and the magnitude and duration of the potential are dependent on the number of neurons involved, the pattern of their discharges and the average strength of postsynaptic currents. 27 Application of Mg2+-free medium removes functional communication blocks between

456

V.V. Senatorov et al.

n e u r o n s a n d thereby facilitates s y n c h r o n i z a t i o n of their discharges, c o n t r i b u t i n g to synchronized seizure activity. Z2'2°'39'47'56I n v o l v e m e n t o f the N M D A receptors in the initiation a n d / o r p r o p a g a t i o n o f epileptiform discharges in M f + - f r e e m e d i u m has been suggested previously. 12'z°'39'47Since the application of N M D A in o u r experiments significantly increased the frequency of SEDs in Mg2+-free A C S F , a n d since the N M D A a n t a g o n i s t s k e t a m i n e a n d A P V completely suppressed all s p o n t a n e o u s epileptiform activity, activation o f N M D A receptors appears to be crucial for the generation of epileptiform activity a n d for its spread across the host graft b o u n d a r y .

CONCLUSIONS O u r present data d e m o n s t r a t e that: (i) the host a n d graft have reciprocal functional c o n n e c t i o n s as s h o w n by recordings of evoked FPs following host or graft stimulation; (ii) these c o n n e c t i o n s could be activated by reducing the extracellular M f + c o n c e n t r a t i o n ; (iii) N M D A application evoked responses in all grafts studied; (iv) b o t h the electrically-evoked FPs a n d the

N M D A - i n i t i a t e d discharges could be inhibited by N M D A antagonists; (v) s p o n t a n e o u s epileptiform discharges developing in Mg2+-free m e d i u m were f o u n d to be s y n c h r o n o u s when c o m p a r i n g graft and host tissues, a n d were mediated by N M D A a n d blocked by its antagonists. O u r findings provide evidence that the N M D A activated r e c e p t o r - c h a n n e l complex contributes to excitatory h o s t - g r a f t synaptic transmission in the neocortex. We would like to thank Prof. Leo P. Renaud for his support during preparation of the manuscript for publication. We are grateful to Dr Charles Yang for his helpful discussion during the writing of the manuscript. We also thank Drs Ralph Nissen, Vance Trudeau, David Mooney and Dan Small for all constructive suggestions stemming from their reading of an earlier version of the manuscript. We also thank Prof. Jan Jansen for his generous gift of a monoclonal antibody against 160,000 mol. wt neurofilament polypeptide. The authors acknowledge the contribution made by anonymous referees in critically analysing the article. This work was supported by grants from the Hungarian Academy of Science. Dr Vladimir Senatorov held a scholarship from the Hungarian Academy of Science. Acknowledgements

REFERENCES

l. Abbes S., Louvel J., Lamarche M. and Pumain R. (1991) Laminar analysis of the origin of the various components of evoked potentials in slices of rat sensorimotor cortex. Electroenceph. clin. Neurophysiol. 80, 310 320. 2. Asher P. and Nowak L. (1987) Electrophysiological studies of NMDA receptors. Trends Neurosci. 10, 284 288. 3. Banczerowski-Pelyhe I., Senatorov V. V., Vil~igi l., Tarnawa I. and Fiil6p Z. (1991) Electrophysiological characterization of neocortical transplants in vitro. Abstr. 3rd 1BRO Congr. Neurosci., Montreal, p. 106. 4. Borroni A. M., Vakhin G., Berry R. and Teyler T. J. (1991) Methods for studying the conductance changes associated with synaptic activation of forebrain slices: the interpretation of field potentials using CGD profiles. J. Neurosci. Meth. 39, 89 102. 5. Bragin A. G., Bohne A. and Vinogradova O. S. (1988) Transplants of the embryonal rat somatosensory neocortex in the barrel field of the adult rat: responses of the grafted neurons to sensory stimulation. Neuroscience 25, 751 758. 6. Burgard E. C. and Hablitz J. J. (1993) Developmental changes in NMDA and non-NMDA receptor-mediated synaptic potentials in rat neocortex. J. Neurophysiol. 69, 230-240. 7. Cambrey-Deakin M. A. and Burgoyne R. D. (1992) Intracellular Ca 2+ and N-methyl-D-aspartate-stimulated neuritogenesis in rat cerebellar granule cell cultures. Devl Brain Res. 66, 25 32. 8. Castro A. J., Sorensen J. C., Tonder N., Bold L. and Zimmer J. (1989) Fetal neocortical transplants grafted into cortical lesion cavities made in newborn rats receive multiple host afferents. A retrograde fluorescent tracer analysis. Restor. Neurol. Neurosci. 1, 13 23. 9. Chagnac-Amitai Y. and Connors B. W. (1989) Horizontal spread of synchronized activity in neocortex and its control by GABA-mediated inhibition. J. Neurophysiol. 61, 747 758. 10. Chervin R. D., Pierce P. A. and Connors B. W. (1988) Periodicity and directionality in the propagation of epiteptiform discharges across neocortex. J. Neurophysiol. 60, 1695 1713. 1l. Cotman C. W., Monaghan D. T., Ottersen O. P. and Storm-Mathisen J. (1987) Anatomical organization of excitatory amino acid receptors and their pathways. Trends Neurosei. 10, 273 280. 12. Dingledine R., McBain C. J. and McNamara J. O. (1990) Excitatory amino acid receptors in epilepsy. Trends pharmac. Sei. 11, 334-338. 13. Floeter M. K. and Jones E. G. (1985) Transplantation of fetal postmitotic neurons to rat cortex: survival, early pathway choices and long-term projections of outgrowing axons. Brain Res. 354, 19 38. 14. Fonseca M., DeFelipe J. and Fair6n A. (1988) Local connections in transplanted and normal cerebral cortex of rats. Expl Brain Res. 69, 387 398. 15. Forsythe I. D., Westbrook G. L. and Mayer M. L. (1988) Modulation of excitatory synaptic transmission by glycin and zinc in cultures of mouse hippocampal neurons. J. Neurosci. 8, 3733-3741. 16. Gardette R., Alvarado-Mallart R. M., Crepel F. and Sotelo C. (1988) Electrophysiological demonstration o f a synaptic integration of transplanted Purkinje cells into the cerebellum of the adult Purkinje cell degeneration mutant mouse. Neuroseience 24, 777-789. 17. Girman S. V. and Golovina I. L. (1990) Electrophysiological properties of embryonic neocortex transplants replacing the primary visual cortex of adult rats. Brain Res. 523, 78 86. 18. Hamasaki T., Komatsu Y . , Yamamoto N., Nakajima S., Hirakawa K. and Toyama K. (1987) Electrophysiological study of synaptic connections between a transplanted lateral geniculate nucleus and the visual cortex of the host rat. Brain Res. 422, 172-177. 19. Harrison N. L. and Simmonds M. A. (1985) Quantitative studies on some antagonists of N-methyl-D-aspartate in slices of rat cerebral cortex. Br. J. Pharmac. 84, 381 39l.

N M D A receptor-mediated graft host connections in the neocortex

457

20. Horne A. L., Harrison N. L., Turner J. P and Simmonds M. A. (1986) Spontaneous paroxysmal activity induced by zero magnesium and bicuculline: suppression by N M D A antagonists and G A B A mimetics. Eur. J. Pharmac. 122, 231 238. 21. Hounsgaard J. and Yarom Y. (1985) Intrinsic control of electroresponsive properties of transplanted m a m m a l i a n brain neurons. Brain Res. 335, 372 376. 22. Huettner J. E. and Bean B. P. (1988) Block of N-methyl-D-aspartate-activated current by the anticonvulsant MK-801: selective binding to open channels. Proc. natn. Aead. Sci. U.S.A. 85, 1307 131 [. 23. Johnson J. W. and Ascher P. (1987) Glycine potentiates the N M D A response in cultured mouse brain neurons. Nature 325, 529 531. 24. Jones K. A. and Baughman R. W. (1988) N M D A - and non-NMDA-receptor components of excitatory synaptic potentials recorded from cells in layer V of rat visual cortex. J. Neurosci. 8, 3522 3534. 25. K e m p J. A., Foster A. C., Leeson P. D., Peiestley T., Tridgett R., lversen L. L. and Woodruff G. N. (19883 7-Chlorokynurenic acid is a selective antagonist at the glycine modulatory sites of the N-methyl-D-aspartate receptor complex. Proc. natn. Aead. Sci. U.S.A. 85, 6547 6550. 26. Kemp J. A., Foster A. C. and Wong E. H. F. (1987) Non-competitive antagonists of excitatory amino acid receptors. Trends Neurosci. 10, 294 298. 27. Langdon R. B. and Sur M. (1990) Components of field potentials evoked by white matter stimulation in isolated slices of primary visual cortex: spatial distributions and synaptic order. J. Neurophysiol. 64, 1484 1501. 28. Lee S. M. and Ebner F. F. (1990) Response characteristics of neocortical graft neurons to host somatosensory input. In Progress in Brain Research (eds Dunnett S. B. and Richards S.-J.). Vol. 82, pp. 301 308. Elsevier Science, Amsterdam. 29. Lester R. A. J. and Jahr C. E. (1992) N M D A channel behavior depends on agonist affinity. J. Neurosci. 12, 635 643. 30. Lodge D. and Johnson K. M. (1990) Noncompetitive excitatory amino acid receptor antagonists. Trends pharmac. Sci. I1, 81 86. 31. LoTurco J. J., Blanton M. G. and Kriegstein A. R. (1991) lnitial expression and endogenous activation of N M D A channels in early neocortical development. J. Neurosci. 11, 792 799. 32. MacDermott A. B. and Dale N. (1987) Receptors, ion channels and synaptic potentials underlying the integrative actions of excitatory amino acids. Trends Neurosci. 10, 2 8 0 284. 33. MacDonald J. F., Miljkovic Z. and Pennefather P. (1987) Use-dependent block of excitatory amino acid currents in cultured neurons by ketamine. J. Neurophysiol. 58, 251 266 34. MacDonald J. F. and Nowak L. M. (1990) Mechanisms of blockade of excitatory amino acid receptor channels. 7)'ends" pharmac. Sei. 11, 167 172. 35. MacDonald J. F. and Wojtowicz J. M. (1980) Two conductance mechanisms activated by applications of L-glutamic, L-aspartic, DL-homocysteic, N-methyl-D-aspartic, and DL-kainic acids to cultured mammalian central neurones. Call. J. Physiol. Pharmae. 58, 1393 1397. 36. Mayer M. L.. Vyklicky L. Jr. and Clements J. (1989) Regulation of N M D A receptor desensitization in mouse hippocampal neurons by glycine. Nature 338, 425 427. 37. Mayer M. L., Westbrook G. L. and Guthrie P. B. (1984) Voltage-dependent block by M f ~ of N M D A responses in spinal cord neurones. Nature 309, 261 263. 38. Mclntyre D. C. and Racine R. J. (1986) Kindling mechanism: current progress on an experimental epilepsy model. Prog. Neurobiol. 27, I 12. 39. Mody I., Lambert J. D. C. and Heinemann U. (1987) Low extracellular magnesium induces epileptilk~rm activity and spreading depression in rat hippocampal slices. J. Neurophysiol. 57, 869 888. 40. Mudrick L. A., Baimbridge, K. G. and Peet, M. J. (1989) Hippocampal neurons transplanted into ischemically lesioned hippocampus: electroresponsiveness and reestablishment of circuitries. Expl Brain Res. 76, 333 342. 41. Naffs E. J., Sorensen J. C., Tonder N. and Castro A. J. (1989) Fetal cortical transplants into neonatal rats respond to thalamic and peripheral stimulation in the adult. An electrophysiological study of single-unit activity. Brain Res. 493, 33 4O. 42. Nowak L., Bregestovski P., Ascher P. Herbet A. and Prochiantz A. (19843 Magnesium gates glutamate-actiw~ted channels in mouse central neurons. Nature 30"7, 462 465. 43. O b u h o v a G. P., Senatorov V. V. and Vartanjan G. A. (1989) Homotopic transplantation of embryonal neocortex tissue into damaged brains of adult rats. Neurosei. behar. Physiol. 19, 324 329. 44. Obuhova G. P.. Gogelia H. K., Senatorov V. V. and F/il6p Z. (1992) Afferent and efferent connections of cortical transplants implanted into the damaged sensorimotor area of the cerebral cortex of adult rats. Neurosei. hehat'. Ph)wiol. 22, 1 5. 45. Peters S., Koh J. and Choi D. W. (1987) Zinc selectively blocks the action of N-methyl-D-aspartate on cortical neurons. Science 236, 589 593. 46. Rashid N. A. and Cambray-Deakin M. A. (1992) N-Methyl-D-aspartate effects on the growth, morphology and cytoskeleton of individual neurons in vitro. Devl Brain Res. 67, 301 308. 47. Schneiderman J. H. and MacDonald J. F. (1989) Excitatory amino acid blockers differentially affect bursting o1" in ~'itro hippocampal neurons in two pharmacological models of epilepsy. Neuroscienee 31, 593 603. 48. Segal M. and Azmitia E. C. (1986) Fetal raphe neurons grafted into the hippocampus develop normal adult physiological properties. Brain Res. 364, 162 166. 49. Senatorov V. V., Nyakas C. and Ffil6p Z. (1993) Visualization of outgrowing axons of grafted neurons by anterograde labelling with Phaseolus ~ulgaris leucoagglutinin in the motor cortex of the rat. Restor. Neurol. Neurosci. 5, 337 345. 50. Senatorov V. V.. O b u h o v a G. P. and Ffil6p Z. (1991) Electrophysiological and morphological properties of embryonic neocortical grafts developing in different regions of the host rat brain. J. neural. Transpl. Plast. 2, 125 140. 51. Senatorov V. V., O b u h o v a G. P. and Silakov V. L. (19913 Influence of nembutal on the impulse activity of cells o9 embryonal nervous tissue transplanted into the rat brain. Neurosci. behat:. Physiol. 21, 311 317. 52. Senatorov V. V., Vilagi I., Tarnawa I., Banczerowski-Pelyhe I. and Ffil6p Z. (19923 Low extracellular magnesium activates functional connections between neocortical graft and surrounding host tissue in rat neocortical slices. Abstr. 4th International Symposium on Neural Transplantation, Washington. Restor. Neurol. Neurosci. 4, 210. 53. Senatorov V. V., Vil~igi 1., Tarnawa I., Banczerowski-Pelyhe I. and Ffil6p Z. (1992) Graft-host glutamatergic neuronal interaction in the neocortex. J. neural Transpl. Plast. 3, 311 312.

458

V . V . Senatorov et al.

54. Shaw C. and Teyler T. J. (1982) The neural circuitry of the neocortex examined in the/n vitro brain slice preparation. Brain Res. 243, 35-47. 55. Tasker J. G., Peacock W. J. and Dudek F. E. (1992) Local synaptic circuits and epileptiform activity in slices of neocortex from children with intractable epilepsy. J. Neurophysiol. 67, 496 507. 56. Vilfigi I., Tarnawa I. and Banczerowski-Pelyhe I. (1991) Changes in seizure activity of the neocortex during the early postnatal development of the rat: an electrophysiological study on slices in MgZ+-free medium. Epilepsy Res. 8, 102 106. 57. Watkins J. C. and Olverman H. J. (1987) Agonists and antagonists for excitatory amino acid receptors. Trends Neurosci. 10, 265-272. 58. Watkins J. C., Krogsgaard-Larsen P. and Honor6 T. (1990) Structure activity relationships in the development of excitatory amino acid receptor agonists and competitive antagonists. Trends pharmac. Sci. l l , 25 33. 59. Wong R. K. S., Miles R. and Traub R. D. (1984) Local circuit interactions in synchronization of cortical neurons. J. exp. Biol. 112, 169 178. 60. Yinon U. and Gelerstein S. (1991) Isochronic transplantation of neonatal grafts in the visual cortex of cats: responsiveness, ocular dominance and specifity of cortical cells to visual stimulation. Expl Brain Res. 87, 181 192. 61. Young A. B. and Fagg G. E. (1990) Excitatory amino acid receptors in the brain: membrane binding and receptor autoradiographic approaches. Trends pharmac. Sci. 11, 126 133. (Accepted 6 June 1994)