- Email: [email protected]
Critical period plasticity of kitten visual cortex is not associated with enhanced susceptibility to electrical kindling
104
Developmental Brain Research, 30 (1986) 1114-10~) Elsevier
BRD 60168
Critical period plasticity of kitten visual cortex is not associated with enhanced susceptibility to electrical kindling MARIA EUGENIA MONETA* and WOLF SINGER Max-Planck-lnstitute for Brain Research, Frankfurt am Main (K R. G. )
(Accepted 24 June 1986) Key words: Kindling - - Visual cortex - - Developmental plasticity - - Kitten
The goal of this study was to determine whether the use-dependent malleability of visual cortex functions which is particularly pronounced in 4-week-old kittens correlates with enhanced susceptibility to kindling. For that purpose the effects of high-frequency electrical stimulation were compared in the visual cortex of 4-week-old kittens and of adult cats. The striate cortex of one hemisphere was stimulated with a single train of pulses whose intensity was set just above the threshold for the elieitation of afterdiseharges (ADs). In kittens the AD thresholds were consistently higher than in adults and with repeated stimulation, the ADs tended to disorganize, to decrease in amplitude and duration and to become more restricted to the site of stimulation after about 6 stimulations. In the adult, by contrast, the ADs remained well organized and constant in duration throughout 30 stimulations. They showed an increase in amplitude and spike frequency and spread with increasing consistency to the other hemisphere. No electrographic or behavioural signs of epileptic activity developed in kittens, while in adults ADs were on occasion followed by irregular spike activity associated with behavioural states resembling absences. We conclude that the visual cortex possesses powerful mechanisms to prevent the development of supracritical excitatory states, these mechanisms being more effective in the kitten than in the adult.
There is now a solid body of data demonstrating use-dependent malleability of cortical functions. O n e line of evidence comes from developmental studies which have revealed that experience can lead to profound modifications of the structural and functional organization of the m a m m a l i a n visual cortex 2'11.
changes in response properties. We wondered therefore, whether these two manifestations of use-dependent neuronal plasticity were based on similar mechanisms. If so one should expect that the developing visual cortex in which experience-dependent malleability is particularly p r o n o u n c e d is more susceptible
A n o t h e r line of evidence for activity-dependent modifiability of cortical networks comes from kindling experiments. Kindling has first been observed in limbic structures 5'6 and refers to the fact that repeated electrical stimulation enhances progressively the excitability of the activated n e u r o n a l structures until generalized seizures develop. Although neocortex 15 and in particular the primary sensory areas turned out to be less susceptible to kindling 1"15than the limbic structures, long-term modifications of cortical excitability were observed with repeated electrical stimulation t3. Both experience-dependent modifications and excitability changes after kindling have in common that neuronal activity entrains long-term
to kindling than the mature cortex which shows much less use-dependent plasticity. In 10 kittens and 5 adult cats stimulation and recording electrodes were implanted in the visual cortex as illustrated in the inset in Fig. 2. The stimulation electrodes consisted of two tapered, teflon-insulated silver wires that were inserted 1 m m deep into the striate cortex with a tip separation of 2 mm. Epidural spherical (1 m m diameter) Ag-AgC1 electrodes served as E E G recording electrodes. The stimuli consisted of a 1-1.5-s train of 0.2-ms pulses delivered at 60 Hz; their intensity ranged from 1.5 to 5 m A and was adjusted individually to evoke a brief afterdischarge ( A D ) at the site of stimulation. Animals were
* Present address: University of Chile, Faculty of Medicine, Department of Physiology and Biophysics, P.O. Box 137-D, Santiago, Chile. Correspondence: W. Singer, Max-Planck-Institute for Brain Research, Deutschordenstrasse 46, D-6000 Frankfurt am Main -71, F.R.G. 0165-3806/86/$03.50 © 1986 Elsevier Science Publishers B.V. (Biomedical Division)
105 divided in two groups: the cats in the first group (6 kittens and 4 adult cats) were stimulated once a day at a fixed time. In all kittens of this group, stimulation was started on postnatal day 28 and continued for two weeks in 4 and for 4 weeks in two kittens. The adult cats received daily stimulation for 4 weeks. In the animals that had been stimulated daily for 4 weeks (two kittens and 4 adults) two stimulations per week were applied thereafter for another 8 weeks to examine the stability of responses. The animals in the second group (4 kittens and 1 adult) were stimulated 3 x a day at intervals not shorter than 3 h for 6 consecutive days, starting at day 28 as in the kittens of the first group. After the end of the experiments all animals were deeply anesthetized with Nembutal and then transcardially perfused with saline followed by 4% formaldehyde. Cryotome serial sections were prepared from the occipital cortex and stained with Thionine for histological examination of the stimulation and recording sites. In kittens, the behavioral response to the kindling stimuli was confined to an occasional disruption of ongoing motor activity and remained unchanged throughout the kindling period. The adult cats responded with contraversive head and body rotations, and two of the adult cats of the first group (1 stimulation per day) developed what appeared like absences. For intervals of 2 - 5 min following stimulation, these cats stopped moving and became unreac-
O
15
k~tten N : 2
~_
_ _ akdtt;sn N~6
9
10
10-
Z
1
2
3
4
5
6
7
8
11
DAYS
12 OF
13
14
15
STIMULATION
Fig. 1. The development of mean AD durations recorded from the stimulation sites in kittens and adults of group 1 during the kindling procedure. Each point represents mean values of AD durations across all subjects for each day of stimulation. Bars indicate S.D.
tive to visual or acoustic stimuli. None of the subjects ever developed a generalized seizure response. The electrographic responses of kittens and adults to high-frequency stimulation differed right from the beginning of the treatment. First, in the kittens the stimulus intensities required for the induction of ADs were about twice as high (3.2 + 0.5 mA) as in the adults (1.5 + 0.3 mA) and thus differed significantly from the latter (P < 0.01, t-test). Second, in the kittens the ADs were confined to the stimulation site and the ipsilateral E E G leads, while in the adults they consistently spread to the contralateral hemisphere. Third, in kittens the frequency of the AD spikes was lower than in adults: in the former the mean frequency was 6.75 ___ 0.96 spikes/s on day 2 while in the latter it was 9 + 0.82 spikes/s. The A D duration in kittens was significantly longer (P < 0.05 in the t-test for paired observations) than in adults throughout the first 6 days of stimulation (Fig. 1), thereafter AD durations in kittens and adults were similar. Further electrophysiological differences between kittens and adults developed during the repeated treatment. First, in the kittens of the first group the duration of the ADs recorded at the site of stimulation decreased until the AD durations became comparable to those observed in the adults. The AD durations in the adults decreased only after the first 3 days and recovered to the initial values by day 8. No further changes in AD duration were observed after the 8th day of treatment neither in kittens nor in adults (Fig. 1). Second, in kittens the amplitude of the AD spikes decreased and the intervals between the spikes became less regular (Fig. 2A), while in adults the AD spikes became larger in amplitude and more regular (Fig. 2B). Third, in all kittens the ADs tended to become restricted to the site of stimulation; after 8 days of stimulation ADs were either no longer recordable (n = 8) or greatly reduced (n -- 2) in amplitude and duration at the ipsi- and the contralateral E E G leads. Fig. 2A illustrates the changes obtained after 12 stimulations. At the site of stimulation the ADs are shorter, less regular and of smaller amplitude than at the beginning of the treatment, and the spread of the A D spikes to regions remote from the stimulated focus has become minimal. In two adult cats, by contrast, the ADs generalized and became more prominent and stable in the recordings from the
106
B
A Day 2
tl ,3-5,
R
(4-6)
F (1-2) ~
Day 5
r~
~
i L (4-6)
1
(1-2)
R (3-5)
(3--5) Day 12
(3-5)
(4--6)
Day 15
L (4-6)
F (1-2)
(1--2)
I100I~V
P
0-
-O
3 A
4 2
Fig. 2. A: electrographic activity elicited in a kitten of the first group after the second and the 12th day of stimulation. On the second day a well-organized AD spreads from the stimulation site (leads 1-2) to the ipsilateral cortex (leads 4-6). Activity does not spread to the contralateral hemisphere (leads 3-5). On day 12, a shorter AD is recorded at the stimulation site. Except for the first 3 s after stimulation, during which some spikes occur at the ipsilateral EEG leads (4-6), there is no further propagation of ADs. B: electrographic responses of an adult cat to high-frequency stimulation at the 5th and the 15th day of stimulation. On the 5th day of stimulation only a small amplitude AD is recorded from the ipsilateral cortex (leads 4-6). No AD occurred over the contralateral cortex (leads 3-5). After the 15th stimulation a well organized and prolonged AD is recorded from the stimulation site (leads 1-2) as well as from the ipsi(4-6) and contralateral leads (3-5). Electrode positions are indicated schematically in the inset below. Stimulation electrodes were located in positions 1 and 2. Bipolar EEG recordings were obtained from positions 4-6 and 3-5 in the ipsi- and contralaterai cortex, respectively.
ipsi- and the contralateral E E G leads (Fig. 2B). In the two other adults of the first group the spatial distribution of the A D s remained unchanged throughout the stimulation period. Fourth, in two adult cats but in none of the kittens electrographic and behavioral abnormalities developed which outlasted considerably the immediate A D triggered by the kindling stimulus. The E E G displayed isolated clusters of sharp waves that occurred in alternation with phases of flattened E E G for up to 20 min. It was during this period that the cats displayed the behavior which appeared to us like absences. Neither the two kittens
nor the adult cats in which daily stimulation was continued for another 3 weeks displayed any further changes in the A D pattern. We therefore discontinued daily stimulation and examined the responses to the kindling stimulus only twice a week for another 8 weeks. Again, we failed to observe any significant changes in the A D properties. In the second group of kittens which had received 3 stimulations per day, A D s initially occurred at the stimulation site and spread to the ipsilateral E E G leads as in the kittens of the first group. From the 10th stimulation onwards A D s had disappeared at
107
R (3-5)
Stim. 2
L (4-6)
~,~,,,..+a,J¢;_- .
F (1-2) "~-: ~+ : - ~
Stirn. 16
f!(3-6) (4-6)
~
[
~
p
,
,
Z , ' r ~ , " ~ .I]IU~I~
%+,-- _ -'l,.,4 :+..~ ~
~'P'~"" ~ ' * ' ~
i-r?. ;.¢ ;" :~_ . "~,,~.,.
(1-2) 1se.....¢
I1oo pv
Fig. 3. Electrographic activity elicited in a kitten of the second group (3 stimulations a day) after the second and the 16th stimulation. After the second stimulation a well-organized AD spreads from the stimulation site (leads 1-2) to the ipsilateral cortex (leads 4-6). Activity as in the first group does not spread to the contralateral hemisphere (leads 3-5). After the 16th stimulation the AD is confined to the stimulation site and shorter and less regular than after the second stimulus. Electrode placement as in Fig. 2.
the ipsilateral E E G leads in all kittens and remained confined to the site of stimulation (Fig. 3). The amplitude and the frequency of A D spikes tended to decrease with increasing number of stimulations. At the site of stimulation A D frequency dropped from a mean of 7.7 _+ 1.8 spikes/s after the second trial to 4.7 + 0.9 spikes/s after the 16th stimulus. In these kittens there was no evidence for a reduction of A D duration throughout the 6 days during which 18 stimulations were applied. Mean A D duration was 12 _+ 4 s on day 1 and 13.1 + 3 s on day 6 and thus the same as in the kittens of the first group at the beginning of the kindling procedure. In the adult eat of group 2 the A D s spread from the site of stimulation to the ipsilateral and in an attenuated way also to the contralateral E E G leads. This pattern did not change significantly throughout 18 trials. In none of the animals did we observe any behavioral signs of seizures. Histological examination of the stimulation sites showed gliosis around the track of the electrodes. There was no evidence for tissue damage other than that caused by the insertion of the stimulation wires. Our results from the adult cats confirm that neo-
cortex 15 and in particular posterior neocortex ~3 has a low susceptibility to kindling. We failed to obtain generalized seizures even though the stimuli elicited well-organized A D s that readily spread to recording sites distant from the site of stimulation. There are only few studies on developmental changes of kindling susceptibility. Since these were performed on subcortical structures and in rat pups, direct comparison with our present results is difficult. Our finding that A D s had higher thresholds in kittens than in adults and spread less readily to remote recording sites agrees with the results of studies in which kindling susceptibility was studied in developing subcortical structures 8'9. In developing rats kindling of the amygdala requires higher stimulus intensities, proceedes at a slower rate and is more lateralized than in adults 4. On the other hand, there is evidence that in the amygdala the refractory period for repeated application of kindling stimuli is shorter in developing than in adult rats I°. Our results are further in agreement with reports which indicate that developing nerve nets show a particularly low responsiveness to high-frequency stimulation 7 and to potentiation of the transcallosal response ~9. Immature excitatory synapses 14 as well as the reduced K+-buffering ability of the still immature glia may be among the causes. What distinguishes our results from those in other developmental investigations of the kindling phenomenon are the changes of the electrographic responses that occurred with repeated stimulation. These changes differed in all aspects from those characteristic for kindling. The following considerations let it appear unlikely that this is due to a growth-dependent dislocation of the stimulation electrodes. First, with the exception of changes of A D duration, the effects of repeated stimulation were similar in the two groups of kittens. This suggests that the relevant parameter was the number of stimulations rather than the time that elapsed between stimuli. Second, there was no change in A D threshold throughout the period of regular stimulation in either group, and even in the two kittens stimulated twice a week for another 8 weeks responses at the site of stimulation remained virtually unchanged. Third, histological examination provided no indication for damage due to shearing between the silver wires and the tissue. Finally, within the relevant time span changes in the dimensions of the occipital pole are small enough to be followed
108 by the very flexible silver wires. By day 28 the dimensions of the visual areas at the occipital pole have reached nearly adult values, much of the remaining growth of the skull being due to expansion in the frontal region and to the differentiation of the frontal sinuses. Another stimulus-independent cause for the deterioration of the ADs and their progressive restriction to the stimulation site could have been ongoing cortical maturation. We consider this possibility as unlikely, too. As mentioned above, the deterioration of ADs depended on the number of stimulations rather than on the increasing age of the kittens. Furthermore, as our results show, ADs were elicitable more easily and spread more readily in the mature than in the immature cortex. Maturation should thus have facilitated rather than antagonized the spread of ADs. Moreover, in the kittens the frequency of regular AD spikes decreased with stimulation while mere age-dependent changes of this parameter should have led to an increase. Only the progressive reduction of AD duration that was observed in the kittens stimulated once a day might be considered as an agedependent phenomenon. In these kittens the AD duration became actually reduced to adult levels after about 8 days of stimulation. These considerations taken together suggest that the progressive attenuation and the spatial restriction of stimulus-induced ADs are actually a consequence of repeated stimulation. This issue contradicts our initial hypothesis and might suggest that experience-dependent malleability and kindling do not depend on the same mechanisms of use-dependent plasticity. There is, however, an alternative interpretation. While it has been shown that the efficacy of excitatory connections can increase in a use-dependent manner during the critical period 7A2As, the bulk of developmental studies emphasize that critical period malleability is characterized by the prevalence of mechanisms capable of
inactivating previously functional connections. One of these mechanisms is activity-dependent competition 16't7. If a cortical cell is activated beyond a critical level 3, excitatory connections that converge onto this cell become weakened if their activity is not sufficiently correlated with that of the postsynaptic cell 16. It is thus conceivable that the strong and poorly structured activation that results from electrical stimulation led to a weakening of excitatory connections that outweighed any concomitant enhancement of excitability. We propose that such a reaction to supracritical and poorly structured global excitation is particularly well adapted to the specific constraints which emerge from use-dependent malleability of excitatory connections in a network which possesses multiple positive feedback loops. Activity-dependent downregulation of excitation such as the inactivation of excitatory connections by competition could serve to keep excitatory interactions below the critical threshold of self-excitation. Since our kittens were not deprived of vision, it is further conceivable that any kindling-dependent increases of excitatory interactions were effectively antagonized by subsequent vision-dependent modifications of excitatory transmission. As is known from experiments with manipulated visual experience, vision-dependent changes of response properties can be induced within a few hours (for rewiew see ref. 2). We suggest that it is because of such developmental mechanisms - which appear to be particularly effective in the visual cortex of 4- to 5-week-old kittens 2 - - that our high-frequency stimuli led to a decrease rather than to an increase of excitability.
1 Baba, H., Facilitatory effects of intermittent photic stimulation on visual cortical kindling, Epilepsia, 23 (1982) 663-670. 2 Fr6gnac, Y. and Imbert, M., Development of neuronal selectivity in primary visual cortex of cat, Physiol. Rev., 64 (1984) 325-434. 3 Geiger, H. and Singer, W., A possible role of Ca++-currents in developmental plasticity, Exp. Brain Res., Suppl.,
in press. 4 Gilbert, M.E. and Cain, D.P., A developmental study of kindling in the rat, Dev. Brain Res.. 2 (1982) 321-328. 5 Goddard, G.V., Mc Intyre, D. and Leech, C.K., A permanent change in brain function resulting from daily electrical stimulation, Exp. Neurol., 25 (1969) 295-330. 6 Goddard, G.V., The kindling model of epilepsy, Trends Neurosci., 6 (1983) 276-279.
We are grateful to Dr. G. Neumann for his technical advice, to Ines Galin for the histological processing and to Gabriele Trauten for typing the manuscript. The travel expenses of M.E. Moneta were partially covered by the International Committee for European Migration (CIME).
109
7 Komatsu, Y., Toyama, K., Maeda, I. and Sakabuchi, H., Long-term potentiation investigated in a slice preparation of striate cortex of young kittens, Neurosci. Lett., 26 (1981) 269-274. 8 Mosh6, S.L., The effect of age on the kindling phenomenon, Dev. Psychobiol., 14 (1981) 75-81. 9 MoshC S.L., Nausie, S.S. and Kaplan, J., Kindling in developing rats: variability of afterdischarge thresholds with age, Brain Res., 211 (1981) 190-195. 10 Mosh6, S.L., Albala, B.J., Ackermann R.F. and Engel, J.,Jr., Increased seizure susceptibility of the immature brain, Dev. Brain Res., 7 (1983) 81-85. 11 Movshon, J.A. and Van Sluyters, R., Visual neural development, Annu. Rev. Psychology, 32 (1981) 477-522. 12 Olson, C.R. and Freeman, R.C., Monocular deprivation and recovery during sensitive period in kittens, J. Neurophysiol., 41 (1978) 65-74. 13 Pinel, J.P., Kindling-induced experimental epilepsy in rat: cortical stimulation, Exp. Neurol., 72 (1981) 559-569.
14 Purpura, P.D., Stability and seizure susceptibility of immature brain. In H. Jasper, A. Ward and A. Pope (Eds.), Basic Mechanisms of the Epilepsies, Churchill, 1969, pp. 481-505. 15 Racine, R., Modification of seizure activity by electrical stimulation: cortical areas, Electroenceph. Clin. Neurophysiol., 38 (1975) 1-12. 16 Rauschecker, J.P. and Singer, W., Changes in the circuitry of the kitten's visual cortex are gated by postsynaptic activity, Nature (London), 280 (1979) 58-60. 17 Wiesel, T.N. and Hubel, D.H., Comparison of the effects of unilateral and bilateral eye closure on cortical unit responses in kittens, J. Neurophysiol., 28 (1965) 1029-1040. 18 Wiesel, T.N. and Hubel, D.H., Extent of recovery from the effects of visual deprivation in kittens, J. Neurophysiol., 28 (1965) 1060-1072. 19 Wilson, D.A. and Racine, R.J., The postnatal development of postactivation potentiation in the rat neocortex, Dew Brain Res., 7 (1983) 271-276.
Developmental Brain Research, 30 (1986) 1114-10~) Elsevier
BRD 60168
Critical period plasticity of kitten visual cortex is not associated with enhanced susceptibility to electrical kindling MARIA EUGENIA MONETA* and WOLF SINGER Max-Planck-lnstitute for Brain Research, Frankfurt am Main (K R. G. )
(Accepted 24 June 1986) Key words: Kindling - - Visual cortex - - Developmental plasticity - - Kitten
The goal of this study was to determine whether the use-dependent malleability of visual cortex functions which is particularly pronounced in 4-week-old kittens correlates with enhanced susceptibility to kindling. For that purpose the effects of high-frequency electrical stimulation were compared in the visual cortex of 4-week-old kittens and of adult cats. The striate cortex of one hemisphere was stimulated with a single train of pulses whose intensity was set just above the threshold for the elieitation of afterdiseharges (ADs). In kittens the AD thresholds were consistently higher than in adults and with repeated stimulation, the ADs tended to disorganize, to decrease in amplitude and duration and to become more restricted to the site of stimulation after about 6 stimulations. In the adult, by contrast, the ADs remained well organized and constant in duration throughout 30 stimulations. They showed an increase in amplitude and spike frequency and spread with increasing consistency to the other hemisphere. No electrographic or behavioural signs of epileptic activity developed in kittens, while in adults ADs were on occasion followed by irregular spike activity associated with behavioural states resembling absences. We conclude that the visual cortex possesses powerful mechanisms to prevent the development of supracritical excitatory states, these mechanisms being more effective in the kitten than in the adult.
There is now a solid body of data demonstrating use-dependent malleability of cortical functions. O n e line of evidence comes from developmental studies which have revealed that experience can lead to profound modifications of the structural and functional organization of the m a m m a l i a n visual cortex 2'11.
changes in response properties. We wondered therefore, whether these two manifestations of use-dependent neuronal plasticity were based on similar mechanisms. If so one should expect that the developing visual cortex in which experience-dependent malleability is particularly p r o n o u n c e d is more susceptible
A n o t h e r line of evidence for activity-dependent modifiability of cortical networks comes from kindling experiments. Kindling has first been observed in limbic structures 5'6 and refers to the fact that repeated electrical stimulation enhances progressively the excitability of the activated n e u r o n a l structures until generalized seizures develop. Although neocortex 15 and in particular the primary sensory areas turned out to be less susceptible to kindling 1"15than the limbic structures, long-term modifications of cortical excitability were observed with repeated electrical stimulation t3. Both experience-dependent modifications and excitability changes after kindling have in common that neuronal activity entrains long-term
to kindling than the mature cortex which shows much less use-dependent plasticity. In 10 kittens and 5 adult cats stimulation and recording electrodes were implanted in the visual cortex as illustrated in the inset in Fig. 2. The stimulation electrodes consisted of two tapered, teflon-insulated silver wires that were inserted 1 m m deep into the striate cortex with a tip separation of 2 mm. Epidural spherical (1 m m diameter) Ag-AgC1 electrodes served as E E G recording electrodes. The stimuli consisted of a 1-1.5-s train of 0.2-ms pulses delivered at 60 Hz; their intensity ranged from 1.5 to 5 m A and was adjusted individually to evoke a brief afterdischarge ( A D ) at the site of stimulation. Animals were
* Present address: University of Chile, Faculty of Medicine, Department of Physiology and Biophysics, P.O. Box 137-D, Santiago, Chile. Correspondence: W. Singer, Max-Planck-Institute for Brain Research, Deutschordenstrasse 46, D-6000 Frankfurt am Main -71, F.R.G. 0165-3806/86/$03.50 © 1986 Elsevier Science Publishers B.V. (Biomedical Division)
105 divided in two groups: the cats in the first group (6 kittens and 4 adult cats) were stimulated once a day at a fixed time. In all kittens of this group, stimulation was started on postnatal day 28 and continued for two weeks in 4 and for 4 weeks in two kittens. The adult cats received daily stimulation for 4 weeks. In the animals that had been stimulated daily for 4 weeks (two kittens and 4 adults) two stimulations per week were applied thereafter for another 8 weeks to examine the stability of responses. The animals in the second group (4 kittens and 1 adult) were stimulated 3 x a day at intervals not shorter than 3 h for 6 consecutive days, starting at day 28 as in the kittens of the first group. After the end of the experiments all animals were deeply anesthetized with Nembutal and then transcardially perfused with saline followed by 4% formaldehyde. Cryotome serial sections were prepared from the occipital cortex and stained with Thionine for histological examination of the stimulation and recording sites. In kittens, the behavioral response to the kindling stimuli was confined to an occasional disruption of ongoing motor activity and remained unchanged throughout the kindling period. The adult cats responded with contraversive head and body rotations, and two of the adult cats of the first group (1 stimulation per day) developed what appeared like absences. For intervals of 2 - 5 min following stimulation, these cats stopped moving and became unreac-
O
15
k~tten N : 2
~_
_ _ akdtt;sn N~6
9
10
10-
Z
1
2
3
4
5
6
7
8
11
DAYS
12 OF
13
14
15
STIMULATION
Fig. 1. The development of mean AD durations recorded from the stimulation sites in kittens and adults of group 1 during the kindling procedure. Each point represents mean values of AD durations across all subjects for each day of stimulation. Bars indicate S.D.
tive to visual or acoustic stimuli. None of the subjects ever developed a generalized seizure response. The electrographic responses of kittens and adults to high-frequency stimulation differed right from the beginning of the treatment. First, in the kittens the stimulus intensities required for the induction of ADs were about twice as high (3.2 + 0.5 mA) as in the adults (1.5 + 0.3 mA) and thus differed significantly from the latter (P < 0.01, t-test). Second, in the kittens the ADs were confined to the stimulation site and the ipsilateral E E G leads, while in the adults they consistently spread to the contralateral hemisphere. Third, in kittens the frequency of the AD spikes was lower than in adults: in the former the mean frequency was 6.75 ___ 0.96 spikes/s on day 2 while in the latter it was 9 + 0.82 spikes/s. The A D duration in kittens was significantly longer (P < 0.05 in the t-test for paired observations) than in adults throughout the first 6 days of stimulation (Fig. 1), thereafter AD durations in kittens and adults were similar. Further electrophysiological differences between kittens and adults developed during the repeated treatment. First, in the kittens of the first group the duration of the ADs recorded at the site of stimulation decreased until the AD durations became comparable to those observed in the adults. The AD durations in the adults decreased only after the first 3 days and recovered to the initial values by day 8. No further changes in AD duration were observed after the 8th day of treatment neither in kittens nor in adults (Fig. 1). Second, in kittens the amplitude of the AD spikes decreased and the intervals between the spikes became less regular (Fig. 2A), while in adults the AD spikes became larger in amplitude and more regular (Fig. 2B). Third, in all kittens the ADs tended to become restricted to the site of stimulation; after 8 days of stimulation ADs were either no longer recordable (n = 8) or greatly reduced (n -- 2) in amplitude and duration at the ipsi- and the contralateral E E G leads. Fig. 2A illustrates the changes obtained after 12 stimulations. At the site of stimulation the ADs are shorter, less regular and of smaller amplitude than at the beginning of the treatment, and the spread of the A D spikes to regions remote from the stimulated focus has become minimal. In two adult cats, by contrast, the ADs generalized and became more prominent and stable in the recordings from the
106
B
A Day 2
tl ,3-5,
R
(4-6)
F (1-2) ~
Day 5
r~
~
i L (4-6)
1
(1-2)
R (3-5)
(3--5) Day 12
(3-5)
(4--6)
Day 15
L (4-6)
F (1-2)
(1--2)
I100I~V
P
0-
-O
3 A
4 2
Fig. 2. A: electrographic activity elicited in a kitten of the first group after the second and the 12th day of stimulation. On the second day a well-organized AD spreads from the stimulation site (leads 1-2) to the ipsilateral cortex (leads 4-6). Activity does not spread to the contralateral hemisphere (leads 3-5). On day 12, a shorter AD is recorded at the stimulation site. Except for the first 3 s after stimulation, during which some spikes occur at the ipsilateral EEG leads (4-6), there is no further propagation of ADs. B: electrographic responses of an adult cat to high-frequency stimulation at the 5th and the 15th day of stimulation. On the 5th day of stimulation only a small amplitude AD is recorded from the ipsilateral cortex (leads 4-6). No AD occurred over the contralateral cortex (leads 3-5). After the 15th stimulation a well organized and prolonged AD is recorded from the stimulation site (leads 1-2) as well as from the ipsi(4-6) and contralateral leads (3-5). Electrode positions are indicated schematically in the inset below. Stimulation electrodes were located in positions 1 and 2. Bipolar EEG recordings were obtained from positions 4-6 and 3-5 in the ipsi- and contralaterai cortex, respectively.
ipsi- and the contralateral E E G leads (Fig. 2B). In the two other adults of the first group the spatial distribution of the A D s remained unchanged throughout the stimulation period. Fourth, in two adult cats but in none of the kittens electrographic and behavioral abnormalities developed which outlasted considerably the immediate A D triggered by the kindling stimulus. The E E G displayed isolated clusters of sharp waves that occurred in alternation with phases of flattened E E G for up to 20 min. It was during this period that the cats displayed the behavior which appeared to us like absences. Neither the two kittens
nor the adult cats in which daily stimulation was continued for another 3 weeks displayed any further changes in the A D pattern. We therefore discontinued daily stimulation and examined the responses to the kindling stimulus only twice a week for another 8 weeks. Again, we failed to observe any significant changes in the A D properties. In the second group of kittens which had received 3 stimulations per day, A D s initially occurred at the stimulation site and spread to the ipsilateral E E G leads as in the kittens of the first group. From the 10th stimulation onwards A D s had disappeared at
107
R (3-5)
Stim. 2
L (4-6)
~,~,,,..+a,J¢;_- .
F (1-2) "~-: ~+ : - ~
Stirn. 16
f!(3-6) (4-6)
~
[
~
p
,
,
Z , ' r ~ , " ~ .I]IU~I~
%+,-- _ -'l,.,4 :+..~ ~
~'P'~"" ~ ' * ' ~
i-r?. ;.¢ ;" :~_ . "~,,~.,.
(1-2) 1se.....¢
I1oo pv
Fig. 3. Electrographic activity elicited in a kitten of the second group (3 stimulations a day) after the second and the 16th stimulation. After the second stimulation a well-organized AD spreads from the stimulation site (leads 1-2) to the ipsilateral cortex (leads 4-6). Activity as in the first group does not spread to the contralateral hemisphere (leads 3-5). After the 16th stimulation the AD is confined to the stimulation site and shorter and less regular than after the second stimulus. Electrode placement as in Fig. 2.
the ipsilateral E E G leads in all kittens and remained confined to the site of stimulation (Fig. 3). The amplitude and the frequency of A D spikes tended to decrease with increasing number of stimulations. At the site of stimulation A D frequency dropped from a mean of 7.7 _+ 1.8 spikes/s after the second trial to 4.7 + 0.9 spikes/s after the 16th stimulus. In these kittens there was no evidence for a reduction of A D duration throughout the 6 days during which 18 stimulations were applied. Mean A D duration was 12 _+ 4 s on day 1 and 13.1 + 3 s on day 6 and thus the same as in the kittens of the first group at the beginning of the kindling procedure. In the adult eat of group 2 the A D s spread from the site of stimulation to the ipsilateral and in an attenuated way also to the contralateral E E G leads. This pattern did not change significantly throughout 18 trials. In none of the animals did we observe any behavioral signs of seizures. Histological examination of the stimulation sites showed gliosis around the track of the electrodes. There was no evidence for tissue damage other than that caused by the insertion of the stimulation wires. Our results from the adult cats confirm that neo-
cortex 15 and in particular posterior neocortex ~3 has a low susceptibility to kindling. We failed to obtain generalized seizures even though the stimuli elicited well-organized A D s that readily spread to recording sites distant from the site of stimulation. There are only few studies on developmental changes of kindling susceptibility. Since these were performed on subcortical structures and in rat pups, direct comparison with our present results is difficult. Our finding that A D s had higher thresholds in kittens than in adults and spread less readily to remote recording sites agrees with the results of studies in which kindling susceptibility was studied in developing subcortical structures 8'9. In developing rats kindling of the amygdala requires higher stimulus intensities, proceedes at a slower rate and is more lateralized than in adults 4. On the other hand, there is evidence that in the amygdala the refractory period for repeated application of kindling stimuli is shorter in developing than in adult rats I°. Our results are further in agreement with reports which indicate that developing nerve nets show a particularly low responsiveness to high-frequency stimulation 7 and to potentiation of the transcallosal response ~9. Immature excitatory synapses 14 as well as the reduced K+-buffering ability of the still immature glia may be among the causes. What distinguishes our results from those in other developmental investigations of the kindling phenomenon are the changes of the electrographic responses that occurred with repeated stimulation. These changes differed in all aspects from those characteristic for kindling. The following considerations let it appear unlikely that this is due to a growth-dependent dislocation of the stimulation electrodes. First, with the exception of changes of A D duration, the effects of repeated stimulation were similar in the two groups of kittens. This suggests that the relevant parameter was the number of stimulations rather than the time that elapsed between stimuli. Second, there was no change in A D threshold throughout the period of regular stimulation in either group, and even in the two kittens stimulated twice a week for another 8 weeks responses at the site of stimulation remained virtually unchanged. Third, histological examination provided no indication for damage due to shearing between the silver wires and the tissue. Finally, within the relevant time span changes in the dimensions of the occipital pole are small enough to be followed
108 by the very flexible silver wires. By day 28 the dimensions of the visual areas at the occipital pole have reached nearly adult values, much of the remaining growth of the skull being due to expansion in the frontal region and to the differentiation of the frontal sinuses. Another stimulus-independent cause for the deterioration of the ADs and their progressive restriction to the stimulation site could have been ongoing cortical maturation. We consider this possibility as unlikely, too. As mentioned above, the deterioration of ADs depended on the number of stimulations rather than on the increasing age of the kittens. Furthermore, as our results show, ADs were elicitable more easily and spread more readily in the mature than in the immature cortex. Maturation should thus have facilitated rather than antagonized the spread of ADs. Moreover, in the kittens the frequency of regular AD spikes decreased with stimulation while mere age-dependent changes of this parameter should have led to an increase. Only the progressive reduction of AD duration that was observed in the kittens stimulated once a day might be considered as an agedependent phenomenon. In these kittens the AD duration became actually reduced to adult levels after about 8 days of stimulation. These considerations taken together suggest that the progressive attenuation and the spatial restriction of stimulus-induced ADs are actually a consequence of repeated stimulation. This issue contradicts our initial hypothesis and might suggest that experience-dependent malleability and kindling do not depend on the same mechanisms of use-dependent plasticity. There is, however, an alternative interpretation. While it has been shown that the efficacy of excitatory connections can increase in a use-dependent manner during the critical period 7A2As, the bulk of developmental studies emphasize that critical period malleability is characterized by the prevalence of mechanisms capable of
inactivating previously functional connections. One of these mechanisms is activity-dependent competition 16't7. If a cortical cell is activated beyond a critical level 3, excitatory connections that converge onto this cell become weakened if their activity is not sufficiently correlated with that of the postsynaptic cell 16. It is thus conceivable that the strong and poorly structured activation that results from electrical stimulation led to a weakening of excitatory connections that outweighed any concomitant enhancement of excitability. We propose that such a reaction to supracritical and poorly structured global excitation is particularly well adapted to the specific constraints which emerge from use-dependent malleability of excitatory connections in a network which possesses multiple positive feedback loops. Activity-dependent downregulation of excitation such as the inactivation of excitatory connections by competition could serve to keep excitatory interactions below the critical threshold of self-excitation. Since our kittens were not deprived of vision, it is further conceivable that any kindling-dependent increases of excitatory interactions were effectively antagonized by subsequent vision-dependent modifications of excitatory transmission. As is known from experiments with manipulated visual experience, vision-dependent changes of response properties can be induced within a few hours (for rewiew see ref. 2). We suggest that it is because of such developmental mechanisms - which appear to be particularly effective in the visual cortex of 4- to 5-week-old kittens 2 - - that our high-frequency stimuli led to a decrease rather than to an increase of excitability.
1 Baba, H., Facilitatory effects of intermittent photic stimulation on visual cortical kindling, Epilepsia, 23 (1982) 663-670. 2 Fr6gnac, Y. and Imbert, M., Development of neuronal selectivity in primary visual cortex of cat, Physiol. Rev., 64 (1984) 325-434. 3 Geiger, H. and Singer, W., A possible role of Ca++-currents in developmental plasticity, Exp. Brain Res., Suppl.,
in press. 4 Gilbert, M.E. and Cain, D.P., A developmental study of kindling in the rat, Dev. Brain Res.. 2 (1982) 321-328. 5 Goddard, G.V., Mc Intyre, D. and Leech, C.K., A permanent change in brain function resulting from daily electrical stimulation, Exp. Neurol., 25 (1969) 295-330. 6 Goddard, G.V., The kindling model of epilepsy, Trends Neurosci., 6 (1983) 276-279.
We are grateful to Dr. G. Neumann for his technical advice, to Ines Galin for the histological processing and to Gabriele Trauten for typing the manuscript. The travel expenses of M.E. Moneta were partially covered by the International Committee for European Migration (CIME).
109
7 Komatsu, Y., Toyama, K., Maeda, I. and Sakabuchi, H., Long-term potentiation investigated in a slice preparation of striate cortex of young kittens, Neurosci. Lett., 26 (1981) 269-274. 8 Mosh6, S.L., The effect of age on the kindling phenomenon, Dev. Psychobiol., 14 (1981) 75-81. 9 MoshC S.L., Nausie, S.S. and Kaplan, J., Kindling in developing rats: variability of afterdischarge thresholds with age, Brain Res., 211 (1981) 190-195. 10 Mosh6, S.L., Albala, B.J., Ackermann R.F. and Engel, J.,Jr., Increased seizure susceptibility of the immature brain, Dev. Brain Res., 7 (1983) 81-85. 11 Movshon, J.A. and Van Sluyters, R., Visual neural development, Annu. Rev. Psychology, 32 (1981) 477-522. 12 Olson, C.R. and Freeman, R.C., Monocular deprivation and recovery during sensitive period in kittens, J. Neurophysiol., 41 (1978) 65-74. 13 Pinel, J.P., Kindling-induced experimental epilepsy in rat: cortical stimulation, Exp. Neurol., 72 (1981) 559-569.
14 Purpura, P.D., Stability and seizure susceptibility of immature brain. In H. Jasper, A. Ward and A. Pope (Eds.), Basic Mechanisms of the Epilepsies, Churchill, 1969, pp. 481-505. 15 Racine, R., Modification of seizure activity by electrical stimulation: cortical areas, Electroenceph. Clin. Neurophysiol., 38 (1975) 1-12. 16 Rauschecker, J.P. and Singer, W., Changes in the circuitry of the kitten's visual cortex are gated by postsynaptic activity, Nature (London), 280 (1979) 58-60. 17 Wiesel, T.N. and Hubel, D.H., Comparison of the effects of unilateral and bilateral eye closure on cortical unit responses in kittens, J. Neurophysiol., 28 (1965) 1029-1040. 18 Wiesel, T.N. and Hubel, D.H., Extent of recovery from the effects of visual deprivation in kittens, J. Neurophysiol., 28 (1965) 1060-1072. 19 Wilson, D.A. and Racine, R.J., The postnatal development of postactivation potentiation in the rat neocortex, Dew Brain Res., 7 (1983) 271-276.