The in vitro neocortical brain slice technique was used to study electrophysio- logical properties of neurons from brain biopsies in 10 patients undergoing neuro- surgical treatment for a variety of conditions, including focal epilepsy.
The principal finding was the occurrence of orthodromically evoked depolariza- tion shifts (DSs) and burst discharges in a proportion of neurons in slices from epileptogenic cortex. These evoked depolarizations and bursts had a number of properties in common with those from experimental epileptogenic foci in neocortex, including large amplitude and prolonged duration; long and variable latencies; and all or none, threshold type behavior, dependent on the parameters of orthodromic stimulation. Also DSs could not be evoked by intracellular stimulation, or blocked by hyperpolarizing current pulses once they had been orthodromically evoked. Responses of DSs to current thus differed markedly from those of neurons in epileptogenie guinea pig hippocampal slices.
The results of these experiments suggest that intracellular events in human neurons involved in epileptogenesis are similar in appearance to those in various animal models. Neurons in chronic epileptogenic foci retain some of their abnormal properties within brain slices maintained in vitro.
INTRODUCTION
Although significant progress has been made recently in studying neuronal mechanisms in acute epileptogenesis by utilizing in vitro mammalian brain slice preparations 3,4,1s-z°,23,ea,2a,27, to date there is a paucity of information about membrane events in neurons of chronic epileptogenic fociS, 14. This is a particularly important issue since the cellular mechanisms involved in acute and chronic epilepto- genesis may differa, 2z. Also, although it has been suggested that neuronal activities in
324
chronic epileptogenesis in primates approximate those which occur in man 22, to date only extracellular recordings have been obtained from human chronic foci2; no intracellular data are available to validate this assumption.
For these reasons, we have applied the in vitro brain slice techniques previously used for hippocampus 21,26 to study neuronal activities in neocortical slices prepared from biopsies of human brain removed during neurosurgical procedures in patients with and without clinical and electrographic manifestations of epilepsy. Our principal objective was to determine whether any of the properties which characterize some neurons in acute or chronic epileptiform foci in vivo, or acute foci in hippocampal or neocortical brain slices in vitro, could be identified in neurons from human epilep- togenic cortex. Preliminary results of these experiments have been published 1~.
METHODS
A total of 10 biopsy specimens were studied. These were derived from temporal (5 cases), frontal (2 cases) and parietal (3 cases) neocortex. Three biopsies were obtained from neocortex of patients who never had epilepsy as a symptom, and one from a patient who had several non-focal seizures 3 years before biopsy. In 6 instances, the clinical history and electroencephalographic data indicated that the patient was having focal seizures involving cortex in the area of the biopsy. This was directly confirmed in 4 patients where electrocorticography was done at the time of surgery and the biopsy taken directly from an area of known epileptogenic discharge. Portions of each biopsy immediately adjacent to the sliced cortex were immersed in formalin and appropriately embedded, sectioned, and stained using the Holtzer method for glial fibers and hematoxylin and eosin for cells.
In all cases, biopsies were obtained from cortex which would ordinarily have been removed at the time of surgery. Informed consent forms approved by the Stanford Medical Committee for Protection of Human Subjects were employed.
In all cases the cortical biopsy was removed at a time when the patient had been anesthetized with agents which potentially have an effect on neuronal properties, such as narcotics and barbiturates. The techniques for slice maintenance, intracellular recording and stimulation, and other manipulations employed in these experiments were similar to those previously described for hippocampal slices z6. Perfusion solutions contained (in mM): NaC1, 124; KCI 5; NaH2PO4 1.25; MgSO4 2; CaCI~ 2; NaHCO3 26; and glucose 10. The pH was 7.4; the osmolality 305 z~ 5 mOsm. Orthodromic stimuli consisted of 0.5-0.05 msec single pulses delivered through monopolar or bipolar sharpened tungsten wires insulated to their tips and placed on the slice ol, the pial or white matter side of the intracellular microelectrode. All experiments were carried out at 37 °C. lntracellular data were accepted for analysis from neurons having membrane potentials o f - -5 5 mV or higher, overshooting spikes of at least 60 mV in amplitude, and input resistances of greater than 20 MfL These neurons generally did not fire spontaneously or show typical injury discharges.
TA
BL
E I
Sum
mar
y o
f cas
e m
ater
ial d
ata
Cas
e no
. D
iagn
osis
B
iops
y si
te
EE
G
Bio
psy
neur
opat
holo
gy
Ort
hodr
omic
bu
rsts
2278
86 (
1)
4191
29 (
2)
5126
85 (
3)
5250
21(4
)
2936
84 (
5)
6109
51 (
6)
5945
61 (
7)
5480
44 (
8)
5663
1911
5 (9
) 63
5121
(10
)
Foc
al s
eizu
res,
tem
pora
l lo
be
Foc
al s
eizu
res,
tem
pora
l lo
be
Tem
pora
l ar
teri
oven
ous
mal
- fo
rmat
ion
(AV
M);
foc
al
tem
pora
l se
izur
es
Rol
andi
c as
troc
ytom
a; f
ocal
m
otor
sei
zure
s G
iant
ane
uris
m m
iddl
e ce
rebr
al a
rter
y tr
ifur
cati
on;
com
plex
par
tial
sei
zure
s T
empo
ral
fibr
illa
ry
astr
ocyt
oma;
foc
al
tem
pora
l se
izur
es
Gli
al c
yst,
ast
rocy
tom
a th
alam
us;
no s
eizu
res;
ag
e 2
year
s P
arie
tal
fibr
illa
ry a
nd
gem
isto
cyti
c as
troc
ytom
a;
3 ge
nera
lize
d se
izur
es,
3 ye
ars
befo
re s
urge
ry
Cys
tice
rcos
is, n
o se
izur
es
Gli
obla
stom
a, p
arie
tal
regi
on;
no s
eizu
res
Sup
erio
r te
mpo
ral
gyru
s F
ocal
spi
kes
on e
lect
ro-
cort
icog
raph
y (E
CG
) of
bi
opsi
ed t
issu
e M
iddl
e te
mpo
ral
gyru
s F
ocal
spi
kes
on E
CG
of
biop
sied
tis
sue
Sup
erio
r te
mpo
ral
Foc
al s
pike
s on
EC
G o
f gy
rus,
adj
acen
t to
bi
opsi
ed t
issu
e A
VM
F
ront
al c
orte
x ad
jace
nt
to t
umor
S
uper
ior
tem
pora
l gy
rus
Ant
erio
r m
iddl
e te
mpo
ral
gyru
s
Par
ieta
l co
rtex
Par
ieta
l co
rtex
Fro
ntop
olar
cor
tex
Par
ieta
l co
rtex
Foc
al s
pike
s on
EC
G o
f bi
opsi
ed t
issu
e B
item
pora
l th
eta
EE
G a
ctiv
ity
wit
hout
epi
lept
ifor
m
pote
ntia
ls
Del
ta f
ocus
, an
teri
or a
nd
mid
-tem
pora
l re
gion
; sp
ike
focu
s, a
nter
ior
tem
pora
l bu
tion
on
EE
G
Non
e
Att
enua
tion
; pa
riet
al a
rea;
fo
cal
slow
ing
tem
pora
l re
gion
; no
spi
kes
Non
e N
one
Nor
mal
cor
tex
+
Nor
mal
cor
tex
+
Rea
ctiv
e gl
iosi
s ÷
Spe
cim
en lo
st
+
Rea
ctiv
e gl
iosi
s +
Pre
serv
atio
n of
cor
tica
l +
arch
itec
ture
wit
h in
vasi
on b
y ab
norm
al
astr
ocyt
es
Nor
mal
cor
tex
0
Gli
osis
0
Nor
mal
cor
tex
0 N
orm
al c
orte
x 0
ta, a
326
A ." ÷-
B - ~ - p - - - - . ~ - A-, : :~. i
C 0.250~
Fig. 1. Intracellular recordings from a neuron which generated membrane depolarizations and burst discharges in response to orthodromic stimuli. A: each stimulus evokes a depolarization and burst at a latency of about 36 msec when interstimulus interval is 800 msec. B: every other stimulus evokes a large depolarization and spike burst when interstimulus interval is decreased to 500 msec. Latency of orthodromically evoked depolarizations is 70 msec. C: responses to first 3 stimuli of B, recorded at higher gain and faster timebase. Small ripples in the baseline following stimuli which do not evoke bursts (second stimulus of C) may represent small EPSPs. Bursts are sometimes followed by after- hyperpolarizations which may last 250 msec or longer (first burst of C). Time calibration in B: 250 msec for A and B. Voltage calibration in C: 50 mV tor A and B; 25 mV for C. Stimuli to cortex near site of impalement. Spikes cut off in C. Membrane potential: --65 mV.
RESULTS
The da ta with respect to the case mater ia l employed are summar ized in Table I.
A to ta l o f 50 in t racel lu lar recordings were ob ta ined f rom slices which had been
prepared f rom biopsies o f cor tex f rom regions thought to be involved in epi lepto-
genesis (cases 1-6). Using ident ical techniques, a popu la t i on o f 20 neurons was
recorded f rom cortex which clinically had no t been involved in epileptogenesis (cases
7-10), and this g roup served as a control* . The pr incipal f inding was the occurrence o f
o r thodromica l ly evoked burs t discharges o f 3 or more spikes in 21 of 50 neurons f rom
epi leptogenic cortex. This cont ras ted with the cont ro l group in which none of 20
neurons showed o r thodromica l ly evoked burs t discharges. Two kinds o f bursts were
noted. In two neurons f rom epi leptogenic cortex, bursts and neurona l behavior closely
resembled that descr ibed for in terneurons in h ippocampus 17 and no rma l neocor tex 16.
Spikes in these neurons had p rominen t fast a f te rhyperpola r iza t ions and there was a
steep re la t ionship between appl ied depolar iz ing current intensity and the frequency o f
spike discharge. A second variety of burst responses had some characteris t ics s imilar to those
previously found in neurons o f acute and chronic neocor t ica l epi leptogenic foci 7,9-11,
la. Responses o f such a neuron f rom epi leptogenic cor tex to o r t hod romic st imuli are
shown in Figs. 1 and 2. Stimuli del ivered at a b o u t 1.25 Hz or slower regularly evoked
* Case 8 is difficult to classify and was included in the control group because no seizures had occurred for 3 years, even off of medication, and the 3 seizures which had occurred had no clear focal components which related them to the biopsied cortex.
327
A , i
B
¢
D
2 3
• m
5 0 msec
Fig. 2. A variety of responses to orthodromic stimuli in neuron of Fig. 1. A: stimuli evoke low amplitude slow depolarizations which may reach threshold to trigger 1 or 2 spikes (A1) or be almost undeteetable in the baseline noise (Aa). B-D: the same intensity stimulus evokes depolarizations lasting 50--100 msec. The burst of B1 is seen in more detail at a faster sweep and higher gain in D. Note the variable duration and amplitude of the depolarizing envelope and pattern of spike discharge in B and C. Time calibration in C1 is 50 msee for segments of A, B and C1. Time calibration in Ca: 50 msee for Ca and Ca. Voltage calibration in C8: 50 mV for A-C.
burst discharges (Fig. 1A); however, at faster rates of stimulation only every other stimulus would evoke a burst (Fig. 1B, C). Stimuli which failed would not necessarily evoke visible EPSPs (Fig. 1B). Small amplitude depolarizing events with a latency about the same as the onset of the expected triggered depolarization and burst were sometimes apparent when burst triggering failed (Fig. 1C, second stimulus). The latency from the stimulus to the onset of the underlying depolarization was long and quite variable. This latency decreased as the stimulus intensity increased and tended to become longer when alternating types of responses were evoked (e.g. latency in Fig. 1A is about 36 msec and in Fig. 1B about 70 msec).
The membrane events and appearance of the burst activity varied considerably from cell to cell and in a single cell. Responses of the neuron of Fig. 1 are shown in more detail in Fig. 2. At a long latency following the orthodromic stimulation, a small ripple in the resting baseline might be seen (Fig. 2Ag). Stimuli on other occasions would evoke slow depolarizations which would trigger anywhere from 2 (A1) to 6 (B1, D) spikes. The underlying depolarizations reached amplitudes of up to 25 mV and were sometimes followed by afterhyperpolarizations which might last 250 msec or longer (Fig. 2B1; Fig. 1B, C). Stimuli which fell during these afterhyperpolarizations were usually not effective in evoking a large depolarizing event and burst response (e.g. Fig. 1C). The spikes evoked by the underlying slow depolarization varied in duration so that the latter spikes in the burst were somewhat broader (e.g. Fig. 2D), how-
328
1
A ..f. ~ 'N. . . , .~
2
" ÷:i ~J I l n A
50msee
Fig. 3. A: orthodromic stimuli (dots) evoke bursts of 3 spikes (Az) riding on a multiphasic depolarizing potential followed by an afterhyperpolarization. As: same stimulus delivered during a hyperpolarizing current pulse which blocks all but one of the burst spikes and uncovers a multiphasic depolarizing event. Sweep is triggered from onset of current pulse. B: responses of the same neuron to increasing intensity current pulses. Arrows in B2 indicate generation of depolarizing afterpotentials (DAPs). Second spike of doublet between arrows may arise from a DAP. Calibrations in B~ for all segments. Upper traces: current monitor.
ever slow spikes similar to those seen in burst discharges in hippocampal pyramidal cells 23 were not detected in the neurons sampled. The highest frequency of spike discharge during bursts was about 160 Hz; spike frequency often decreased after the initial 2-3 spikes (Fig. 2D). No stereotyped 'long first interval '2,22 bursts were seen in the population sampled. In marked contrast to previous findings in chronic foci in monkey neocortex 14, or in acute foci in neocortex s,l° or hippocampal slices 19, no clearcut spontaneous bursting, or prominent epileptiform field potentials were seen. This is however also characteristic of acute epileptogenesis produced by penicillin in neocortical slices of guinea pig (Prince, Wong and Gutnick, unpublished).
Another neuron which generated bursts of 3-4 spikes following orthodromic stimuli is shown in Fig. 3. When the orthodromic stimulus was delivered during hyperpolarizing current pulses, the underlying depolarizing event could be uncovered (Fig. 3A2). More intense hyperpolarizing pulses blocked all spikes and revealed a complex depolarizing potential which had a number of inflections. The amplitude of the underlying depolarizing envelope increased as the membrane potential was increased (see below and Fig. 5). In the bursting neuron of Fig. 3, depolarizing current pulses evoked trains of spikes (B1) and intense depolarization produced what appeared to be depolarizing afterpotentials (DAPs) following some spikes (Fig. 3B2, arrows) and an occasional 'extra' spike which appeared to arise from a DAP (second spike of doublet between arrows in Fig. 3B2). Similar DAPs were noted in neurons from non- epileptogenic cortex. In no instance did depolarizing pulses evoke self-sustained burst discharges, as is the case in hippocampal CA3 pyramidal neurons 23.
In several neurons, although no burst was evoked by orthodromic stimuli, a long latency, large amplitude depolarizing event followed a short latency EPSP (Fig. 4A2, B2-3). The membrane potential of neocortical neurons in vitro tends to be quite high and spontaneous activity is unusual except following injury (Wong and Prince, unpublished observations). Because of the high resting membrane potentials, depol-
A
1 2 3 4
1
329
- llnA
125m v 50msec
Fig. 4. Effects of changes in stimulus intensity (A) and stimulus frequency (B) on EPSP and longer latency depolarizing event in a neuron from epileptogenic cortex which generates only a single spike during large depolarizations. At: stimulus evokes isolated EPSP. As: more intense stimulus evokes a larger EPSP followed by long latency depolarizing event with an amplitude of almost 25 mV. As: stronger orthodromic stimulus evokes a larger EPSP separated by inflection from the second depolariz- ing event which is evoked at shorter latency. A4: additional increase in stimulus intensity results in almost complete fusion between fast rising short latency EPSP and larger depolarizing event. BI: response similar to A4 seen at higher gain and faster sweep. B2: 6th response to a train of ortbodromic stimuli delivered at 800 msec intervals. First response of the train resembled B1. Ba: 6th and 7th re- sponses to a train of stimuli delivered at 500 msee intervals. Interstimulus interval changed from 800 to 500 msec between B2 and B~. B4: 6th and 7th responses to train of stimuli delivered at 250 msec intervals, when frequency was changed after recording B3. Calibrations in A4:50 msec, 25 mV for sweeps of A. Calibrations in B4 for sweeps of B. Spike amputated in B1, blocked in B~-4. Upper trace: current moni- tor. Orthodromic stimuli delivered during hyperpolarizing current pulses in B~-4.
arizing events with amplitudes of up to 25 mV might trigger only a single spike without
evoking a burst (Fig. 4A4). These long latency depolarizing potentials decreased in latency without significantly changing in amplitude as the stimulus intensity was
increased (Fig. 4A2-4). The long latency depolarizations tended to appear in an all or
none manner at a threshold stimulus intensity (cf. Figs. 4A1 and A2). As noted above and in Fig. 1, increasing the stimulus frequency had an effect on the latency and security of triggering the large amplitude depolarizing events. For example, in Fig. 4B stimuli at 0.5 Hz (B1) regularly evoked a depolarization reaching 25 mV. The effects of increasing stimulus frequency on this response are shown in Fig. 4B2-a. The response
of B2 was evoked at the end of a train of stimuli delivered every 800 msec. The short latency EPSP is uncovered as a prominent inflection on the rising phase of the late
large amplitude depolarization, and the peak latency and onset of the latter event are delayed as stimuli frequency is increased (cf. Fig. B1, B2 and B3). Between B2 and B3 the stimulus interval was decreased from 800 to 500 msec. As is shown in B3, at this
stimulus frequency the late depolarizing event has decreased in amplitude, and increased in peak latency; stimuli alternately evoked EPSPs or EPSPs followed by late depolarizations. During a train of stimuli delivered at 250 msec intervals the late depolarization fails altogether, uncovering an isolated EPSP (B4).
In 5 neurons, it was possible to assess the effects of hyperpolarizing and depolarizing intracellular current pulses of up to 1.5 nA on the depolarizing event underlying burst discharge. A typical result is seen in Fig. 5. This neuron was the only uninjured one in which we recorded what appeared to be a single spontaneous burst
330
A
50msec
Fig. 5. A: spontaneous depolarization and burst discharge. B: responses of the neuron in A to ortho- dromic stimuli (dots) at resting potential (--70 mV) (1 st frame) and at increasing membrane potentials produced by hyperpolarizing current pulses (2nd and 3rd frames). Time calibration in A: 50 msec. Spikes cut off in all segments.
discharge (Fig. 5A); however this could have been related to an undetected mechanical effect since the micromanipulator had been touched a few seconds before to adjust the impalement. In any case, in this neuron orthodromic stimuli evoked typical depolari- zations and bursts (Fig. 5B, 1st frame) whose latency was not significantly affected during hyperpolarizing current pulses (second and third frames of B). The underlying depolarizing envelope increased progressively in amplitude as the membrane was hyperpolarized (B, 2nd frame) and the rate of rise of this potential also increased with hyperpolarization. With sufficient current it was possible to block the spike discharges completely, leaving a depolarizing event of about 50 mV. Close inspection of this potential revealed that it contained a number of inflections, especially on its peak and falling phases (B, 3rd frame). Depolarizing current pulses decreased the amplitude of the orthodromically evoked depolarizing potential (not shown), but, as noted above, self-sustained bursts and intrinsic depolarization shifts could not be evoked in neocortical neurons using currents of up to 1.5 nA.
In contrast to the burst generation demonstrated in a portion of the neuronal population from epileptogenic cortex described above, other neurons in these slices or in slices from 'normal' cortex (Table I, cases 7-10) generated only EPSPs or EPSP-IPSP sequences and one or two spikes following orthodromic stimulation. No long latency, large amplitude depolarizing events resembling those of Figs. 1-5 were noted in these cells. A more detailed description of the membrane properties and other behavior of neurons in normal human cortical biopsies is in preparation (Prince, Wong and Basbaum, in preparation, and ref. 16).
DISCUSSION
The principal finding of these experiments is that some neurons from neocortical slices of patients with epilepsy tend to generate prolonged depolarizations and associated bursts of spikes following orthodromic stimulation. The first question to be raised is whether this really represents abnormal behavior which would allow us to label such neurons as 'epileptic', or whether there is a subpopulation of neocortical neurons which generate similar bursts under 'normal' conditions. For example, in hippocampal slices spontaneous or directly evoked (but not orthodromically evoked - - see refs. 24 and 25) burst discharges occur in CA3 pyramidal neurons 2a. The only
331
parameter which distinguishes normal CA3 pyramidal cell burst behavior from that which is regarded as epileptogenic (e.g. following application of penicillin, strychnine or other convulsants), is the presence of a repetitive multiphasic extracellular field potential indicative of synchronization of a population of neurons. This extracellular potential appears to be equivalent to the field potential 'spikes' seen in hippocampus in vivo (see ref. 18 for discussion).
In the absence of such an 'index' response, it becomes difficult to relate the behavior of a single cell to epileptogenesis. Unfortunately, prominent spontaneous or evoked epileptiform field potentials are not a feature of epileptogenesis in guinea pig neocortical slices even after penicillin treatment (Prince and Wong, unpublished observations). Exposure of such slices to penicillin nonetheless induces prominent orthodromically evoked depolarization shifts and burst generation in most neurons; such burst activities are, in other respects, identical to those occurring in vivo (Prince, Gutnick, and Wong, in preparation). From this it follows that the absence of spontaneous burst discharges or orthodromically evoked multipeaked field potentials is not evidence against the epileptogenic nature of the bursting behavior described in some neurons from human neocortex in these experiments. The relatively small number of biopsies studied to date does make it difficult to conclude that certain cell characteristics are specific for epileptogenesis. It would be important to collect data from a control group of neurons in injured, but nonepileptogenic human neocortex to determine if burst generating neurons are present under such conditions. Ideally, electrocorticography should be done in this control tissue to rule out the presence of EEG epileptiform events which might not have been detectable in scalp recordings. Practical and ethical considerations will make it difficult to obtain this type of control data.
The possibility that burst generation in neurons from epileptogenic cortex is only a reflection of some adverse response to the slicing and incubating procedures, rather than the usual abnormal behavior of these neurons as they existed in vivo must also be considered. Against this interpretation are the findings that burst generating neurons generally had high resting membrane potentials, overshooting spikes, and the other criteria of 'health' which have been applied to intracellular recordings from slices in our laboratories. Also, uninjured neurons which generate orthodromic depolariza- tions and bursts similar to those of Figs. 1-5 are not a common feature in normal human or guinea pig neocortex in vitro*.
The morphology of bursts in neurons of slices from human epileptogenic cortex and those seen in neurons recorded in chronic alumina cream neocortical foci in vivo is similar (cf. spike bursts of Figs. 1, 2 and 5 with those in Figs. 5 and 6 of ref. 14). Qualitative similarities in the appearances of slow depolarizations and spikes of course do not necessarily support a common mode of burst generation. Certain characteristics noted in human neocortical burst generating neurons are also similar to those of
* We have recorded rare neurons in such normal slices which generate either bursts on direct depolari- zation, or at short latency foUowing orthodromic stimulation. This behavior may disappear as the impalement improves, and hyperpolarizing the cell membrane eliminates burst generation.
332
neurons in penicillin foci of neocortex studied in vivo 1,7-12. For example, in both cases the long but variable latency for evoking the depolarization shift is markedly sensitive to stimulus intensity and frequency, but large latency shifts may occur with little effect upon the amplitude and form of the underlying depolarization shift. Also sustained depolarizations and associated burst generation cannot be evoked by depolarizing current pulses. The slow depolarization underlying bursts can be increased and decreased by hyperpolarizing and depolarizing intracellular current pulses respec- tively, however it is not possible to block underlying depolarizations with strong hyperpolarizing current even when all the spikes are blocked. These characteristics also closely match those obtained in penicillin-treated neocortical slices (Prince, Gutnick and Wong, unpublished), but are strikingly different from those reported for hippocampal pyramidal cells in slices bathed with the same drug where depolarization shifts evoked by orthodromic stimuli may be blocked by intracellular hyperpolariza- tion, and bursts may be directly evoked by intracellular depolarizationslS-20, z4. It is not known whether such differences in behavior reflect fundamental differences in synaptic connectivity and neuronal structure, or variations in the activities and distribution of ionic conductances over the neuronal membranes of neocortical versus hippocampal neurons (see ref. 13 for discussion).
In summary, there is a population of neurons in slices from chronic epilepto- genic cortex of humans which exhibits long and variable latency depolarizations and multiple spike burst discharges following orthodromic activation. These neurons have a number of properties in common with those from experimental epileptogenic foci in neocortex. If this behavior is regarded as characteristic of neurons involved in epileptogenesis, two important conclusions follow. First, the findings directly confirm that the intracellular events in human neurons involved in epileptogenesis are similar in appearance to those of various animal models used to date. Secondly, it appears that neurons in chronic epileptogenic foci retain some of their abnormal properties within brain slices maintained in vitro. This will provide an opportunity to examine cellular mechanisms of chronic focal epileptogenesis in greater detail in future studies.
ACKNOWLEDGEMENTS
These experiments were supported by NIH Grants NS 06477 and NS 12151 from the NINCDS. We thank Dr. Barry Tharp who performed electrocorticography and interpreted EEGs, and Drs. J. Hanbery, G. Silverberg, F. Conley and R. Britt of the Division of Neurosurgery for their cooperation in obtaining the cortical biopsies. We are also indebted to Dr. Lucien Rubinstein and his staff who performed histological studies on the biopsy material and Dr. Philip Schwartzkroin who helped in initial experiments.
REFERENCES
1 Ayala, G. F., Dichter, M., Gumnit, R. J., Matsumoto, H. and Spencer, W. A., Genesis of epileptic interictal spikes: new knowledge of cortical feedback systems suggests a neurophysiological ex- planation of brief paroxysms, Brain Research, 52 (1973) 1-17.
2 Calvin, W. H., Ojemann, G. A, and Ward, A. A., Jr., Human cortical neurons in epileptogenic loci:
333
comparison of interictal firing patterns to those of 'epileptic' neurons in animals, Electroenceph. clin. NeurophysioL, 34 (1973) 337-351.
3 Courtney, K. R. and Prince, D. A., Epileptogenesis in neocortical slices, Brain Research, 127 (1977) 191-196.
4 Dingledine, R. and Gjerstad, L., Penicillin blocks hippocampal IPSPs, unmasking prolonged EPSPs, Brain Research, 168 (1979) 205-209.
5 G16tzner, F. L., Fetz, E. E. and Ward, A. A., Jr., Neuronal activity in the chronic and acute epilep- togenic focus, Exp. Neurol., 42 (1974) 502-518.
6 Jasper, H. H., Application of experimental models to human epilepsy. In D. P. Purpura, J. K. Penry, D. Tower, D. M. Woodbury and R. Walter (Eds.), Experimental Models of Epilepsy, Raven Press, New York, 1972, pp. 505-602.
7 Matsumoto, H., Intracellular events during the activation of cortical epileptiform discharges, Eleetroenceph. Clin. Neurophysiol., 17 (1964) 294-307.
8 Matsumoto, I-I. and Ajmone-Marsan, C., Cortical cellular phenomena in experimental epilepsy: interictal manifestations, Exp. NeuroL, 9 (1964) 286--304.
9 Matsumoto, H., Ayala, G. F. and Gumnit, R. J., Neuronal behavior and triggering mechanisms in cortical epileptic focus, J. Neurophysiol., 32 (1969) 688-703.
11 Prince, D. A., The depolarization shift in 'epileptic' neurons, Exp. Neurol., 21 (1968) 467~,85. 12 Prince, D. A., Cellular activities in focal epilepsy. In M. A. B. Brazier and F. Coceani (Eds.), Brain
Dysfunction in Infantile Febrile Convulsions, Raven Press, New York, 1976, pp. 187-212. 13 Prince, D. A., Neuronal events underlying epileptogenesis. In R. A. Ouvrier and P. G. Procopis
(Eds.), Proc. 2nd International Child Neurology Congress, Spectrum Publications, in press. 14 Prince, D. A. and Futamachi, K. J., Intracellular recordings from chronic epileptogenic foci in the
monkey, Electroenceph. clin. Neurophysiol., 29 (1970) 496-510. 15 Prince, D. A. and Wong, R. K. S., Electrophysiology of human epileptic neurons, Neurosci. Abstr.,
5 (1979) 196 (abstract 647). 16 Prince, D. A., Wong, R. K. S. and Basbaum, A. I., Membrane properties ofidentified human cortical
neurons, Neurosci. Abstr., 4 (1978) 79 (Abstract 227). 17 Schwartzkroin, P. A. and Mathers, L. H., Physiological and morphological identification of a non-
pyramidal hippocampal cell type, Brain Research, 157 (1978) 1-10. 18 Schwartzkroin, P. A. and Prince, D. A., Penicillin-induced epileptiform activity in the hippocampal
in vitro preparation, Ann. Neurol., 1 (1977) 463~,69. 19 Schwartzkroin, P. A. and Prince, D. A., Cellular and field potential properties of epileptogenic
hippocampal slices, Brain Research, 147 (1978) 117-130. 20 Schwartzkroin, P. A. and Prince, D. A., Changes in excitatory and inhibitory synaptic potentials
leading to epileptogenic activity, Brain Research, 183 (1980) 61-76. 21 Skrede, K. K. and Westgaard, R. I-[., The transverse hippocampal slice: a well-defined cortical
structure maintained in vitro, Brain Research, 35 (1971) 589-593. 22 Ward, A. A., Jr., The epileptic neuron: chronic foci in animals and man. In I-I. H. Jasper, A. A. Ward,
Jr. and A. Pope (Eds.), Basic Mechanisms of the Epilepsies, Little, Brown, Boston, 1969, pp. 263-288. 23 Wong, R. K. S. and Prince, D. A., Participation of calcium spikes during intrinsic burst firing in
hippocampal neurons, Brain Research, 159 (1978) 385-390. 24 Wong, R. K. S. and Prince, D. A., Dendritic mechanism underlying penicillin induced epileptiform
activity, Science, 204 (1979) 1228-1231. 25 Wong, R. K. S., Prince, D. A. and Basbaum, A. I., Intradendritic recordings from hippocampal
neurons, Proc. nat. Acad. Sci. (Wash.), 76 (1979) 986-990. 26 Yamamoto, C., Intracellular study of seizure-like afterdischarges elicited in thin hippocampal
sections in vitro, Exp. NeuroL, 35 (1972) 154-164. 27 Yamamoto, C. and Kawai, N., Seizure discharges evoked in vitro in thin section from guinea pig
