Developmental Brain Research 123 (2000) 151–164
www.elsevier.com / locate / bres
Research report
Activity-dependent development of spontaneous bioelectric activity in
organotypic cultures of rat occipital cortex
´ 1 , Klaus Albus*
Diego Echevarrıa
¨
, Germany
Department of Neurobiology /192, Max Planck Institute for Biophysical Chemistry, P.O. Box 2841, D-37070 Gottingen
Accepted 2 August 2000
Abstract
The development of spontaneous bioelectric activity (SBA) in organotypic tissue cultures (OTCs) from rat occipital cortex was studied
by means of extracellular recording techniques in OTCs grown normally for 6–51 days in vitro (DIV), and in OTCs in which SBA had
been silenced from DIV 4 on for 2 to 3 weeks by elevating the Mg 21 levels in the growth medium. The proportions of spontaneously
active neurones increased from about 25% at 6–14 DIV to more than 80% beyond the third week in vitro. Mature neurones discharged at
shorter intervals and more vigorously than immature neurones; the developmental increase in firing rate was not significant, however. In
OTCs 6–14 DIV the majority of spontaneously active neurones fired sluggishly in a regular manner. The remaining neurones fired action
potentials in the form of discrete bursts resembling interictal activity in vivo. The proportions of these neurones increased from about 40%
at 6–14 DIV to more than 80% beyond the third week in vitro. During development in vitro the mean burst duration increased from 3.5 s
to about 8 s whereas the mean burst rate (between 0.7–1 bursts / min) remained constant. Activity-deprived neurons had low firing rates
and fired action potentials in the form of discrete bursts with a mean burst rate of 0.4 / min. The proportions of spontaneously active
neurons, the variability of neuronal firing and the viability of the explants either were not altered by the activity blockade or had recovered
to control values after 5–6 days in normal growth medium. We conclude that in OTCs of rat neocortex the absence of SBA during
development in vitro delays the maturation of excitatory mechanisms responsible for the developmental increase in firing intensity. The
development of burst firing modes is less affected by activity blockade. 2000 Elsevier Science B.V. All rights reserved.
Theme: Development and regeneration
Topic: Cerebral cortex and limbic system
Keywords: Neocortex; Spontaneous action potential; Extracellular recording; Explant culture; Activity blockade
1. Introduction
The development of structure and function of the
mammalian visual system is conditional upon the presence
of spontaneous bioelectrical activity. Continuously silencing retinal activity blocks the structural development of
retinogeniculate synapses [31], delays the growth of
*Corresponding author. Tel.: 149-551-201-1652; fax: 149-551-2011788.
E-mail address: kalbus@gwdg.de (K. Albus).
1
Present address: Department of Morphological Sciences, Faculty of
Medicine, 30071 Murcia, Spain.
geniculocortical axon arbors [2] and of synaptogenesis in
the visual cortex [47] and interferes with the development
of ocular domains in the superior colliculus [51] and the
visual cortex [50]. Even before the onset of visual experience, spontaneous retinal activity seems to be necessary for the formation of highly stereotyped patterns of
connections in the lateral geniculate nucleus (LGN) [40].
Impulse blockade in the LGN has been shown to retard the
developmental expression of a broad range of receptors in
the developing visual cortex [26]; intracranial infusion of
the sodium channel blocker tetrodotoxin (TTX) impairs the
initial targeting made by thalamic axons when projecting
to the visual cortex [12].
The effect of impulse blockade in central regions of the
visual system on the development of bioelectric activity in
these structures has not been investigated so far in vivo.
0165-3806 / 00 / $ – see front matter 2000 Elsevier Science B.V. All rights reserved.
PII: S0165-3806( 00 )00089-4
152
´ , K. Albus / Developmental Brain Research 123 (2000) 151 – 164
D. Echevarrıa
Blocking neural activity in the retina is not a means of
exploring the activity-dependent development of bioelectric activity in the visual cortex since total visual deafferentation does not block spontaneous activity in visual
cortical neurons [33]; the remaining activity seems to
guarantee a normal development of visual responsiveness
of cortical neurons even after longer periods of binocular
impulse blockade [50]. Thus, in order to study the activitydependent development of structure and function in central
regions of the nervous system, long-term in vitro systems
like dissociated cell cultures or organotypic cultures
(OTCs) have been employed.
Neurons in explant cultures of spinal cord and neocortex
generate spontaneous bioelectric activity (SBA) in the
absence of stimulation [16,13]. Crain et al. [17] demonstrated that characteristic spontaneous and evoked synaptic
network discharges could be recorded from embryonic
mouse cerebral and spinal cord tissues explanted at presynaptic stages and maintained for many weeks in culture
media containing impulse-blocking concentrations of
xylocaine or high Mg 21 prior to return to control medium
(for review see Ref. [20]). Others have demonstrated that
the development of SBA is strongly influenced by impulse
blockade starting shortly after explantation or cell seeding.
The directions of the effects observed are not however,
uniform and seem to depend on the type of culture and the
brain region cultivated. SBA is abnormally low in OTCs of
spinal cord which had been exposed to TTX [5,6]. In
contrast, following the cessation of TTX-treatment a
prolonged increase in electrical excitability of the neuronal
networks in OTCs of cerebellum [49] and dissociated cell
cultures from neocortex [14,42] occurs. The increases in
mean firing rate of single neurons and in proportions of
neurons with stereotyped bursting have been accounted to
a decrease in GABA neurotransmission [45]. The suggestion of an increase in the ratio of excitatory to inhibitory mechanisms in activity-deprived cultures is corroborated by findings that TTX treatment of OTCs reduces
GAD-mRNA expression (glutamic acid decarboxylase; the
synthetic enzyme for GABA) in neocortical interneurones
[46] and significantly reduces the number of somatic
inhibitory synapses on Purkinje cells [49]. The increased
frequency of calcium transients in activity-deprived dissociated cell cultures from neocortex also indicates hyperactivity [38].
We have investigated the development of SBA in OTCs
from rat occipital cortex in normally growing OTCs and in
OTCs in which SBA was continuously blocked for 2–3
weeks. We show that activity blockade significantly reduces the level of SBA in neocortical OTCs and in
addition, alters the temporal organization of spike trains.
Thus, OTCs from neocortex react differently to activity
blockade than dissociated cell cultures from neocortex or
OTCs from cerebellum. Parts of the findings have been
published in abstract form [22].
2. Materials and methods
2.1. Preparation of organotypic tissue cultures
Organotypic brain cultures were prepared and maintained in vitro using the roller-tube technique introduced
by Hogue [29] and later modified by Costero and Pomerat
¨
[15] employing the protocol given by Gahwiler
[25].
Briefly, rats (Wistar strain) 2 or 3 days old were anaesthetised by lowering their body temperature on ice and
decapitated, and one cortical hemisphere was removed.
The cortex was dissected according to the atlas of Paxinos
et al. [39]; visual areas in very young rats cannot be
distinguished histologically. Blocks of occipital cortex
were cut parasagittally into 350-mm-thick slices on a
McIlwain tissue chopper. Slices were placed on a Petri
dish with Gey’s balanced salt solution (GBSS) supplemented with D-glucose to a final concentration of 0.65%
at 48C for a 30–60-min recovery period. Individual slices
were then placed on cleaned coverslips (11324 mm), fixed
by a drop of chicken plasma coagulated with thrombin (20
ml each) and placed into reagent tubes supplied with 0.75
ml semi-artificial medium (50% Eagle’s basal medium,
25% Hank’s balanced salt solution (HBSS), 25% of
inactivated horse serum, 1 mM final concentration of
L-glutamine, 0.65% final concentration of D-glucose and
1.26 mM final concentration Ca 21 ). All procedures were
done in sterile conditions. Slices were incubated at 368C in
dry air up to 12 weeks in a roller tube drum (10 rev. / h).
After 2–3 days an antimitotic cocktail (5-fluoro-2-desoxyuridine, cytosine-bD-arabinofuranoside and uridine; each
at 4 mM final concentration) was added to the nutrient
medium for 24 h. Thereafter, medium was changed every
3–4 days. Gross histological structure of the slices was
revealed by Nissl staining.
2.2. Data acquisition and evaluation
Cultures from 6 to 51 DIV were placed in a recording
chamber and continuously superfused with HBSS at a rate
of 25 ml / h (bath temperature 36618C); in all experiments,
the final concentration of Ca 21 in the superfusion medium
was increased to 3.1 mM. The pH in the recording
chamber was adjusted to the pH in the test tube in which
the explant was grown and was held constant within 60.2.
In addition, the air in the recording chamber was saturated
with oxygen. Extracellular recordings were performed with
glass capillaries (1.5-mm inner tip diameter) filled with
superfusion medium.
Spike trains of spontaneously active neurones were
stored on magnetic tape and analysed off-line using
software commercially available (Brain-Wave) or developed by Dr. Gras (R-Time; Dept. of Zoology, Universi¨
ty Gottingen).
Neuronal activity was analysed by calculating interspike interval (ISI) distributions and autocorrelog-
´ , K. Albus / Developmental Brain Research 123 (2000) 151 – 164
D. Echevarrıa
rams (ACG). The following parameters were measured for
each spike train: firing rate (number of spikes / s), median
value and modal value of ISI distributions, the position
(latency) of peaks in ISI histograms, width and latency of
peaks in ACG histograms, duration and rate (number / min)
of grouped spike discharges (denoted in the following as
bursts; see Ref. [28]) and burst ratio. Burst duration was
measured either directly from the spike time histograms
(STHs; binwidth 1 s) or from ACG histograms (binwidth
50 ms). The burst ratio was determined by calculating the
number of intervals shorter than the mean interval as a
proportion of all interspike intervals, following the procedure proposed by Habets et al. [28] (see also Ref. [42]).
In addition to burst ratio we note the burst ratio standardised with respect to firing rate. The variability of
spontaneous activity over time was determined by calculating the coefficient of variation of the firing rate in a sliding
window 120 s long (CV120s) which was moved in time
steps of 60 s. ACGs were subjected to a x 2 -test for identity
with a homogeneous distribution. ACGs with peaks significantly above the expected value for a homogeneous
distribution (P value ,0.001) were differentiated from
those without significant peaks.
Single units were classified into two types: non-spontaneously active neurones (NonSp-neurones) and spontaneously active neurones (Sp-neurones). Sp-neurones with a
burst ratio ,5 were denoted as spontaneously nonbursting
neurones (SpnonB-neurones) and Sp-neurones with a burst
ratio $5 as spontaneously active bursting neurones (SpBneurones). The proportion of Sp-neurones was calculated
on the basis of all neurones recorded per developmental
stage. Proportions of SpB-neurones per developmental
stage were calculated considering only neurones which
could be recorded for at least 4 min. NonSp-neurones were
detected by electrically stimulating the gray or white
matter at various distances from the recording site with a
glass insulated tungsten electrode (stimulation frequency
0.2 Hz, pulse width 50 ms).
Neurones firing spontaneously five action potentials or
less within a period of 10 min (N587) were classified as
rarely spontaneous active neurones and included into the
group of NonSp-neurones. A total of 19 single units with
modal intervals of $5000 ms (N58) and / or with a median
value in the ISI distribution exceeding 20 s (N511), which
were present at all developmental stages, were accepted as
spontaneously active but were excluded from calculation
of mean values and medians.
The normal development of spontaneous bioelectric
activity (SBA) in vitro was studied in 41 OTCs; 39 OTCs
were explanted at PND 2, and two OTCs at PND 3. From
these 41 OTCs 579 neurones were recorded. For quantitative analysis we considered only SP-neurons recorded
longer than 4 min (N5260). The experiments were run
longitudinally: OTCs from one hemisphere (batch) were
used at different developmental stages. The number of
153
OTCs from the same batch contributing to one stage was
#2 (6–7 DIV, 15–21 DIV, 22–28 DIV, 29–35 DIV) or #3
(8–14 DIV, .35 DIV). Analysis of variance (see below) of
the intra-stage data demonstrated no difference between
the OTCs contributing to one stage. The few exceptions
found (P value ,0.05; less than 2% of all comparisons)
did not contribute to inter-stage differences if present. We
therefore, considered the data points at different developmental stages to be independent.
In another set of experiments SBA was chronically
blocked in OTCs for 14–23 days by elevating the Mg 21
levels in the growth medium; blockade started at DIV 4.
The blockade is due to the lowered Ca 21 / Mg 21 ratio
which blocks synaptic transmission [21,43]. High levels of
Mg 21 do not necessarily block spontaneous discharges in
cerebellar explant cultures [24,49,53]. We therefore, tested
the effectiveness of 5 mM and 11 mM Mg 21 , respectively,
in acute experiments in three explants (22–28 DIV, 65
neurones). The chronic experiments were performed in 14
OTCs which had been returned to normal medium for 0–6
days. Since the normal development study had revealed
that the increase in the proportion of Sp-neurones and the
improvement of firing properties were prominent at the end
of the third postnatal week, the activity blockade was
induced at 4 DIV and continued at least until DIV 18 and
at the most until DIV 27. After that time, OTCs were
returned to normal medium and explored with microelectrodes. From the 285 neurones recorded, 76 Sp-neurones
were quantitatively analysed as described above. Recordings at recovery day 0 (Rec(0)) started 3 h after returning
to normal Mg 21 levels. As controls (control (s.h.)) four
OTCs (65 neurones; 52 quantitatively analysed) from the
hemispheres from which the explants for the chronic
activity blockade experiments were taken, were grown
under normal conditions. The data from these controls
were also considered for the analysis of normal development.
Differences in mean values between developmental
stages were calculated by using the Kruskal Wallis test
with Dunn’s multiple comparison test as post test. Differences were accepted as statistically significant if the P
value (two-tailed) was less than 0.05 (GraphPad Prism, V
2.0).
3. Results
3.1. Normal development of SBA in vitro
3.1.1. Active sites and firing variability
The experimental approach consisted in searching for
spontaneous and / or electrically evoked activity of single
units by moving the electrode through the tissue in a
direction nearly perpendicular to the surface. If a neurone
was detected the site was denoted as an active site. In the
154
´ , K. Albus / Developmental Brain Research 123 (2000) 151 – 164
D. Echevarrıa
normal development study at least 14 sites per OTC were
explored (mean6S.D.518.862.7); an OTC was considered for evaluation only, if at least nine neurones could be
recorded. The mean number of neurones recorded / OTC
was 14.264.5; the proportions of active sites / sites explored per OTC range between 20% and 100% and are on
average 75.6616.7% (mean6S.D.; Table 1). A statistical
analysis demonstrates that OTCs recorded from shortly
after explantation have about as many active sites as OTCs
at later developmental stages.
The variability of SBA over time was determined for
each Sp-neurone by calculating the coefficient of variation
of the firing rate (CV120s; Table 1). The CV120s ranges
from 11.68 to 204.3; the range of single values and the
median of the distributions do not change during development in vitro. Thus, the neural populations recorded at
different developmental stages are homogeneous with
respect to the parameter firing variability over time.
3.1.2. Developmental changes in firing strength and
temporal structure of spike trains
The main developmental events during time in vitro are
increases in the proportions of Sp-neurones and SpBneurones and increases in the firing intensity of Sp-neurones. During the first and second week in vitro only about
25% of the neurones are spontaneously active (Table 1).
Spike trains characteristic of these early developmental
stages are reproduced in Fig. 1. About 60% of the Spneurones fire sluggishly at a rate of usually less than 0.2
Hz (Fig. 1A–C) without any clear indication of discrete
bursts (Fig. 1C9). The remaining neurones fire action
potentials in the form of short discrete bursts, at rather long
intervals during which either irregular spontaneous activity
(Fig. 1D) or no spontaneous activity (Fig. 1E) occurs.
Thus, 37.5% of these neurones were classified as SpBneurones at 6–7 DIV, and 18.7% at 8–14 DIV. Bursts last
from about 200 ms to s; the intraburst firing rate seldomly
exceeds 20 Hz. In 80% of the neurones one type of burst is
repeated at similar intervals with rather constant duration
and intraburst firing rate (Fig. 1E). In the remaining 20%
the repetition rate is irregular or two different types of
bursts are produced (see below, Fig. 2).
The proportions of Sp-neurones increase to about 60%
in the third week and to more than 80% at later stages of
development in vitro. At later developmental stages most
of the Sp-neurones are classified as SpB-neurones (Table
1); accordingly, the proportions of neurones displaying
regular nonburst activities drop below 20%. Beyond the
third week in vitro the majority of neurones now fire in the
form of discrete bursts at rather long intervals. Bursts last
on average longer (7–8 s) than during the first 2 weeks in
vitro (3–4 s; see Table 5) whereas the burst rate has not
changed (Table 2). Typical spike trains recorded in OTCs
3 weeks and longer in vitro are presented in Fig. 2. Often a
burst continues over 10 s; in a few cases (Fig. 2C and E)
bursts last up to 30 s. The intraburst firing rate often
exceeds 20 Hz. Bursts have either a predominantly tonic
(Fig. 2B, B9; D, D9; E, E9) or phasic-tonic (Fig. 2A, A9; C,
C9) appearance; tonic discharges may be interrupted by
short silent intervals (Fig. 2B, B9; C, C9; E, E9). The
neurone in Fig. 2E9 in fact, fires repetitive short bursts at a
rate of |3 Hz (see also Figs. 5 and 6). In more than 80% of
these neurones stereotyped burst patterns consisting of one,
or more than one burst type are seen. For example, three of
the four discharge sequences of the neurone shown in Fig.
2C consist of an initial burst which after a short silent
period is followed by a long tonic discharge lasting up to
30 s. The discharge sequence of another neurone (Fig. 2D)
starts with an initial burst lasting about 5 s which after a
short pause is followed by a 10–15 s long burst and
subsequently by a number (N58–10) of shorter bursts; this
pattern is repeated with minor variations.
The development of SBA in individual batches of OTCs
(i.e. OTCs explanted from one hemisphere) with respect to
the time course of development, and to proportions of
Sp-neurones (Fig. 3A–D) matches closely the mean developmental trend shown in Table 1; i.e. interbatch variability is generally low. The delayed increase in the
proportion of Sp-neurones in one OTC as shown in Fig. 3F
was an exception.
3.1.3. Quantitative evaluation of developmental changes
in SBA
SBA was quantified by the firing rate, the median and
the modal value of the ISI distribution, the burst ratio, the
standardised burst ratio and the burst rate (Table 2). Two
different developmental trends are apparent. From the first
to the second week in vitro the firing intensity decreases
Table 1
Development of SBA in OTCs of rat occipital cortex. Summary of methodical and basic experimental data. Average recording time (all neurons) was
11.463.4 min (mean6S.D.)
Days in vitro
6–7
8–14
15–21
22–28
29–35
.35
No. batches
No. OTCs
No. neurons
Active sites (%; mean6S.D.)
CV120s (mean6S.D.)
Sp-neurons (%)
SpB-neurons (%)
5
7
87
67.3624.5
49.8621.5
26.4
37.5
3
8
112
83.2613.5
58.2643.3
25
18.7
5
8
119
80.2619.5
51.2636.7
59.7
59.6
4
8
98
66.5615.1
62.4637.5
87.7
81.5
4
5
88
7767.3
67.1635.5
85.2
89.5
2
5
75
77.962.8
55.4631.4
82.7
81.8
´ , K. Albus / Developmental Brain Research 123 (2000) 151 – 164
D. Echevarrıa
Fig. 1. Spontaneous activity of single neurons in OTCs of rat occipital
cortex 6–14 DIV. (A–E) Spike time histograms (binwidth 1 s, duration 10
min). Underlined segments of C, D and E are shown at an enlarged time
scale in C9, D9 and E9 (binwidth 40 ms, duration 25 s). The scale at the
left hand side of each histogram indicates the number of action potentials /
bin. All OTCs were explanted at PND 2. The table summarizes the firing
characteristics of the neurons
6–14 DIV
A
B
C
D
E
DIV
Firing rate (Hz)
Median of ISI (ms)
Mode of ISI (ms)
Burst ratio
Burst rate (bursts / min)
7
0.19
3369
1449
2.1
–
6
0.14
5761
51
1.2
–
14
0.18
2450
25
1.6
–
7
0.27
534
20
2.8
–
7
0.66
51
27
5.3
2.3
and neurons fire at larger intervals; in addition, the burst
ratio decreases. The differences are not significant, however. The proportions of Sp-neurones remain low and
fewer neurons are classified as SpB-neurones at 8–14 DIV
than at 6–7 DIV. A spike train typically seen at the end of
the second week in vitro is shown in Fig. 1C.
155
Fig. 2. Spontaneous activity of single neurons in OTCs of rat occipital
cortex $15 DIV. (A–F) Spike time histograms (binwidth 1 s, duration 10
min). Underlined segments of A–F are shown at an enlarged time scale in
A9–F9 (binwidth 40 ms, duration 25 s). The scale at the left hand side of
each histogram indicates the number of action potentials / bin. All OTCs
were explanted at PND 2. All neurons shown were classified as SpBneurons. The table summarizes the firing characteristics of the neurons
15–51 DIV
A
B
C
D
E
F
DIV
Firing rate (Hz)
Median of ISI (ms)
Mode of ISI (ms)
Burst ratio
Burst rate (bursts / min)
51
1.21
53
15
27.1
0.3
41
2.1
109
72
15.5
4.8
51
3.73
48
46
49.7
0.4
21
0.68
194
77
25.7
2.3
36
1.53
54
22
33.2
0.5
41
0.59
126
34
5.9
3.5
Beginning with the third week in vitro the firing
intensity progressively increases and stable values are
reached during the fourth week in vitro. Neurons fire more
vigorously and at shorter intervals. Most neurons fire in the
burst mode as indicated by the significant increases in
burst ratio beyond the third week in vitro; these increases
are independent from increases in firing rate. The mean
firing rate doubles from the first week to stages beyond the
third week although due to the large scatter in values the
increase is not significant (P.0.05). Interestingly, the burst
´ , K. Albus / Developmental Brain Research 123 (2000) 151 – 164
D. Echevarrıa
156
Table 2
Developmental changes in firing intensity and temporal structure of spike trains in OTCs of rat occipital cortex a
Days in vitro
6–7
8–14
15–21
22–28
29–35
.35
No. batches
No. OTCs
No. neurons
Firing rate (Hz)
4
5
16
0.34
(0.21–1.25)
242
(125–677)
105
(36–174)
3
(1.7–5.7)
8.6
(3.2–10.8)
2
4
16
0.13
(0.09–0.23)
2904
(723–5135)
30
(23–440)
1.8
(1.6–3.4)
14.1
(9.8–24.8)
5
7
52
0.30
(0.17–0.80)
189
(74–544)
53
(21–112)
6.7
(2.6–18.3)
19.2
(10.9–34.6)
1.01
(0.70–1.72)
4
6
65
0.63
(0.24–1.28)
66(2)
(23–60)
26(2)
(20–48)
17.2(1)
(7.2–27.9)
28.7(1)
(17–49.3)
0.67
(0.46–1.08)
4
5
67
0.5
(0.24–1.67)
89
(61–164)
30(2)
(23–60)
16.4(1)
(9.8–70.3)
37.7(1)
(27.2–56.9)
0.75
(0.44–1.13)
2
5
44
0.63
(0.26–1.77)
116
(54–189)
38
(15–51)
10.3(1)
(6.1–24.2)
18.8(1)
(10.5–47)
0.77
(0.52–1.76)
Median of ISI (ms)
Mode of ISI (ms)
Burst ratio
Burst ratio / firing rate
Burst rate (burst / min)
0.74
(0.41–2.1)
a
The numbers for firing rate, median value and modal value of ISI distribution, burst ratio, burst ratio / firing rate and burst rate give the median and (in
parentheses) the 25th and 75th percentile of the respective distributions. Median values from stages beyond 8 DIV significantly higher or lower (P value
,0.05) than median values of data from OTCs 6–7 DIV are marked with (1) or (2), respectively.
rate does not change during time in vitro. This indicates
that in spite of the increase in firing intensity the timing of
bursts present shortly after explantation is maintained at
later developmental stages in vitro. Further changes in
firing behavior concern burst duration and the internal
structure of bursts (see below).
3.2. Development of SBA in activity-deprived OTCs
3.2.1. Acute effects of elevating Mg 21 concentration
The effects of elevating the Mg 21 concentration on SBA
of single neurons were tested in three OTCs in which 65
neurons were recorded. Each OTC was studied for about 8
Fig. 3. Development of spontaneous activity in OTCs of rat occipital cortex. (A–D) Intra- and interbatch variability in the proportion of Sp-neurones
(white bars) and NonSp-neurons (black bars) during development in vitro. Each histogram represents recordings made at different DIV in OTCs explanted
from one hemisphere. All OTCs were explanted at PND 2. Numbers above the histograms indicate numbers of recorded neurons per OTC; numbers
underneath histograms indicate DIV.
´ , K. Albus / Developmental Brain Research 123 (2000) 151 – 164
D. Echevarrıa
h and the average number of sites explored per OTC was
26. As a control 10 randomly distributed sites were
explored for SBA before changing the Mg 21 concentration. Then magnesium levels were elevated in the
superfusion medium and after waiting for about 1 h in two
of the three explants the effects of this manipulation on
SBA were investigated by exploring at least 10 sites / OTC.
Application of superfusion medium containing 5 mM or 11
mM magnesium had completely abolished SBA in these
cases. After returning to normal superfusion medium the
explants were again explored with microelectrodes and
Sp-neurones were found in normal proportions. However,
the activity of these neurons as judged on the basis of
average firing rate, median value and modal value of ISI
distribution and burst ratio was lower than in age-matched
controls and did not recover within 2 h. In the third explant
therefore, the time course of the effects of 11 mM Mg 21
were studied in more detail. About 20 randomly distributed
sites were explored within 15–45 min after elevating the
magnesium level. A few Sp-neurones were detected 15–30
min, and no SBA was found 30–45 min after starting with
11 mM Mg 21 . After returning to normal superfusion
medium SBA recovered to about normal levels within 3 h.
From these findings we concluded that 5 mM or 11 mM
magnesium in the superfusion medium effectively block
SBA in OTCs of rat occipital cortex. In order to reliably
block SBA over longer periods OTCs were grown with 11
mM.
3.2.2. Active sites and firing variability
SBA in activity-deprived OTCs was tested at recovery
day zero (Rec(0)) and up to 6 days after returning to
normal concentrations of Mg 21 in the growth medium
(Rec(1) to Rec(6)). At Rec(0) recording was started after 3
157
h of superfusing the explant with normal Mg 21 concentrations.
Methodical and experimental data for the recovery
period are summarized in Table 3. The proportions of
active sites / sites explored are on average 84.4613.6%
(mean6S.D.). Comparisons with age-matched cases from
the normal development population (22–28 DIV; Table 1)
reveal slightly higher proportions of active sites in the
deprived OTCs.
The differences in variability of SBA over time
(CV120s; Table 3) during the recovery period (Table 3)
and between pooled recovery data and control cases and
age-matched controls (Table 4) were not significant.
3.2.3. SBA and temporal organisation of spike trains in
activity-deprived OTCs
The activity blockade induces a number of abnormal
firing characteristics, which distinguish deprived from
normally grown explants. Throughout the recovery period
activity-deprived neurons discharge sluggishly at low mean
rates. The differences in mean firing rates and median
interspike interval values between recovery data and data
from controls (s.h.) and age-matched controls, respectively
(Table 4), are significant. In contrast, modal values of ISI
distributions are not different between recovery data and
control cases. This assumes that activity-deprived neurons
fire their rarely occurring action potentials in form of short
bursts at long intervals. In fact, standardizing burst ratio
with respect to firing rate reveals that the proportion of
neurons discharging at shorter intervals is significantly
higher in the activity-deprived cases than in the controls
(Tables 3 and 4). A further persistent abnormal property of
activity-deprived OTCs is a burst rate only about half of
that present in the controls (Tables 3 and 4). On the other
Table 3
SBA in activity-deprived OTCs of rat occipital cortex (experiment 2)a
Recovery stage
Rec(0)
Rec(1–4)
Rec(5–6)
Days in vitro
No. batches
No. OTCs
No. neurons
Active sites (%; mean6S.D.)
CV120s (mean6S.D.)
Sp-neurons (%)
SpB-neurons (%)
Firing rate (Hz)
Median of ISI (ms)
Mode of ISI (ms)
Burst ratio
Burst ratio / firing rate
Burst rate (burst / min)
18–27
3
9
173
80.3615
78.5644.4
33.1
100
0.12 (0.07–0.24)
383 (160–841)
28 (7–487)
12.4 (8.7–18.2)
108 (23–168)
0.45 (0.24–0.8)
20–27
3
3
72
95.465
102.4650.4
47.2
90
0.12 (0.11–0.22)
437 (120–1022)
9 (4–610)
11 (8.1–18.8)
82 (33–169)
0.35 (0.26–0.45)
28–29
2
2
40
8961.4
68.5640.5
90
86.2
0.18 (0.1–0.34)
471 (346–623)
296 (18–491)
12.3 (8.9–17.7)
84 (32–143)
0.49 (0.37–0.72)
a
Data from the recovery period (0 to 6 days) was grouped into Rec(0), Rec(1–4) and Rec(5–6). Average recording time (all neurons) was 10.764.1 min
(mean6S.D.). For the quantitative analysis only spontaneous active units were included which were recorded longer than 4 min (Rec(0) N527, Rec(1–4)
N520 and Rec(5–6) N529). The differences in activity parameters between recovery periods were not significant. For further information see footnote to
Table 2.
158
´ , K. Albus / Developmental Brain Research 123 (2000) 151 – 164
D. Echevarrıa
Table 4
Effects of activity blockade on the development of SBA in OTCs of rat occipital cortex a
Pooled recovery data
Control s.h.
Days in vitro
(18–29)
(20–30)
Age-matched
normal controls
(22–35)
No. neurons
CV120s (mean6S.D.)
Firing rate (Hz)
Median of ISI (ms)
Mode of ISI (ms)
Burst ratio
Burst ratio / firing rate
SpB-neurons (N)
Burst rate (burst / min)
76
81646
0.1 (0.14–0.24)
421 (316–637)
80.2 (7.5–495.1)
11.6 (8.6–18.5)
89 (48–147)
70
0.4 (0.3–0.7)
52
65636.9
0.5 (0.28–1.22) (1)
89 (61–149) (2)
30 (24–54)
16.8 (9.9–36.8) (1)
35 (23–57) (2)
59
0.7 (0.4–1.2) (1)
132
64.8636.5
0.6 (0.2–1.3) (1)
78 (44–172) (2)
27.3 (19.5–50)
14 (5.7–24.5) (1)
32 (20–50.3) (2)
113
0.7 (0.4–1.1) (1)
a
Data from the recovery period following activity blockade were pooled and compared with control s.h. data and data from normally grown age-matched
explants. Data from control (s.h.) cases were not different from age-matched normal controls (P value .0.05). Median values from control (s.h.) cases and
age-matched controls significantly higher or lower (P value ,0.05) than median values of pooled recovery data are marked with (1) and (2), respectively.
Control s.h., OTCs from the same hemispheres providing the explants for the chronic activity blockade experiments.
hand, the mean burst duration is about the same as
measured in age-matched controls (Table 5).
Other abnormalities in firing behavior in activity-deprived OTCs disappear during the recovery period (Table
3). The proportions of Sp-neurones are 33% at Rec(0) and
thus resemble proportions seen during the first and second
week of normal development (see Table 1). Later in the
recovery period the proportions of Sp-neurones rapidly
increase and reach normal levels at Rec(5–6). The high
proportions of SpB-neurones recorded at Rec(0) decreased
to normal levels (about 80%) at Rec(5–6).
Typical examples of the temporal organization of spike
trains in activity-deprived OTCs are documented in Fig. 4.
In many respects the STHs compiled with a binwidth of 1
s (Fig. 4A–F) resemble the STHs seen in OTCs normally
grown for 2–3 weeks. The neurones fire action potentials
in the form of discrete bursts, albeit at intervals longer than
in the controls. Between bursts either irregular sparse
spontaneous activity (Fig. 4C, D, F) or no spontaneous
activity at all (Fig. 4A, B, E) occurs. Bursts have either a
tonic (Fig. 4A, D, F) or a phasic-tonic appearance (Fig.
4B, C, E) and they last from about 200 ms to 20 s; the
intraburst firing rate is lower than in control cases and
seldomly exceeds 10 Hz. In 90% of the spike trains with
discrete burst patterns one type of burst with rather
constant duration and intraburst firing rate is repeated at
similar intervals. Spike trains with two or more different
burst types as observed during later stages of normal
development in vitro (see Fig. 2D) were not found.
Striking differences between activity-deprived neurones
and normally grown age-matched neurones are noted if
bursts are analysed at an enlarged time scale (binwidth 40
ms; Fig. 4A9–F9). In normally developing neurones (Fig.
2A9–F9) often a burst consists of a short phasic component
followed by a longer lasting tonic discharge. In activitydeprived neurones a longer ‘tonic’ burst component consists of a number of phasic subbursts usually lasting less
than 1 s which are repeated at short, regular intervals (Fig.
4A9, C9, D9, F9). The firing rate during a subburst may
exceed 50 Hz which is similar to firing rates measured
during the initial phasic burst components in normally
grown neurones (Fig. 2A9, C9). The number of neurones
firing in the ‘phasic subburst’ mode decreases during
recovery from activity blockade.
Table 5
Firing characteristics of single neurons in activity-deprived (recovery) and normally developing OTCs of rat occipital cortex a
Latency 1 peak ISI (ms)
Latency 2 peak ISI (ms)
Width central peak ACG (ms)
Latency central peak ACG (ms)
Burst duration (s)
Recovery
6–14 DIV
15–28 DIV
.28 DIV
12 (8–29)
380 (120–600)
200 (60–950)
20 (10–30)
8 (6–11)
28 (16–115) (1)
140 (52–340)
390 (180–1000)
30 (27.5–65) (1)
3.5 (0.95–10)
32 (20–50) (1)
96 (60–185) (2)
1000 (650–1250) (1)
50 (30–67.5) (1)
8 (5–12.5)
36 (20–48) (1)
88 (48–165) (2)
1100 (800–1400) (1)
50 (30–100) (1)
7.5 (5–13)
a
Normal development data were grouped into 6–14 DIV, 15–28 DIV and .28 DIV (comprising data from OTCs 29–51 DIV). The numbers give the
median and (in parentheses) the 25th and 75th percentile of the distribution per stage. The latency of peaks in ISI distributions and of the central peak in
ACGs correspond to the position of the bin at peak maximum. The width of the central peak in ACGs (binwidth 10 ms, time difference axis 3 s) is
measured as the difference between lagtime zero and the lagtime at which the peak fell below the expected value for a homogeneous distribution. The burst
duration is determined only for SpB-neurons; it is measured either directly from the spike time histograms (binwidth 1s) or from ACGs with a binwidth of
50 ms or 100 ms (time difference axis 15 s and 30 s, respectively). The number of neurons per group is .50 with exception of group 6–14 DIV, in which
22 ISI distributions and eight SpB-neurons were evaluated. Median values of normal development data significantly higher or lower (P value 0.05) than
median values of recovery data are marked with (1) and (2), respectively. Firing characteristics did not change during normal development with the
exception of the width of the central ACG peak which increased significantly from 6–14 DIV to .28 DIV.
´ , K. Albus / Developmental Brain Research 123 (2000) 151 – 164
D. Echevarrıa
Fig. 4. SBA in activity-deprived OTCs of rat occipital cortex. (A–F)
Spike time histograms (binwidth 1 s, duration 10 min). Underlined
segments of A–F are shown at an enlarged time scale in A9–F9 (binwidth
40 ms, duration 25 s). The scale at the left hand side of each histogram
indicates the number of action potentials / bin. All OTCs were explanted at
PND 2. All neurons shown were classified as SpB-neurons. The table
summarizes the firing characteristics of the neurons
Recovery
A
B
C
D
E
F
DIV
Recovery day
Firing rate (Hz)
Median of ISI (ms)
Mode of ISI (ms)
Burst ratio
Burst rate (bursts / min)
28
4
0.26
691
640
13.8
0.7
24
0
0.17
332
10
8.4
0.6
20
0
0.21
69
6
15.3
0.2
25
2
0.19
62
6
5.4
0.2
29
6
0.41
165
18
22.8
1.1
28
5
1.83
33
7
12.3
0.6
3.3. Temporal characteristics of spike trains: a
comparison between activity-deprived and normally
grown OTCs
The temporal organisation of spike trains was further
investigated by analysing interspike interval distributions
(ISIs; Fig. 5) and autocorrelogram histograms (ACGs; Fig.
6) in more detail.
The developmental increase in firing intensity of normally grown neurones (see Table 2) is reflected in the ISI
159
distributions (Fig. 5; 6–14 DIV, 15–51 DIV). At 6–14
DIV many neurones often display multiple peaks in their
ISIs, thus firing at intervals shorter than 100 ms and in
addition, at intervals longer than 200 ms. At later developmental stages the proportions of longer intervals and
accordingly, the number of secondary and tertiary peaks in
the ISI distributions are reduced; in addition the spikes
sampled over a 10-min period have increased in numbers
(Fig. 5, 15–51 DIV). ISI distributions in activity-deprived
OTCs, in particular from recovery days 0–3 (Fig. 5,
Recovery) have a short latency peak followed after at least
200 ms by distinct second and third peaks; these distributions resemble distributions recorded during the first
and second week in vitro. However, in activity-deprived
OTCs the latency of the first peak is shorter and that of the
second peak longer than in normally developing OTCs
(Table 5). This type of distribution reflects the ‘phasic
subburst’ firing mode described above (Fig. 49, C9, D9, F9).
ACG histograms in the activity-deprived cases resemble
that seen in normally developing OTCs 6–14 DIV in
normal development (Fig. 6). In particular, proportions of
ACGs without a significant peak (Fig. 6, 6–14 DIV, A; see
Section 2) are similar between both groups (35% and 38%,
respectively); in the normal development cases longer than
14 DIV these proportions account to less than 15%. ACGs
with significant peaks which in addition to a tonic, slowly
decaying flank display a distinct initial phasic component
(Fig. 6: Recovery (E, F); 6–14 DIV (D, E); 15–51 DIV (C,
E, F)), account to 39% in both activity-deprived cases and
normal developing cases 6–14 DIV. At later stages of
normal development these proportions exceed 59%. Tonic
ACGs with a slowly decaying flank lacking a distinct
initial phasic component (Fig. 6: Recovery (A), 15–51
DIV (A, B, D)) increase in proportions during normal
development (from 20 to 28%) and are rare in the activitydeprived cases (10%) in which they are seen only at
recovery days 4–6. Proportions of ACGs (Fig. 6: Recovery
(B–D), 6–14 DIV (B, C)) dominated by a short initial
phasic component preferentially occur during early phases
of recovery from deprivation and during 6–14 DIV of
normal development.
The quantitative evaluation of ACG data (Table 5)
supplements the conclusions made from analysing ISI
distributions. The width of the central ACG peak is smaller
and the latency of the central ACG peak shorter in activitydeprived neurones than in normally developing neurones.
The differences most likely reflect the differences in the
temporal organisation of bursts: many activity-deprived
neurones discharge in the phasic ‘subburst’ mode, whereas
normally grown neurones prefer phasic-tonic or tonic burst
discharges.
We occasionally observed neurones oscillating at stable
frequencies (see Fig. 6; Recovery (F), 15–51 DIV (E))
ranging between 0.7 and 50 Hz. The proportions of these
neurones were 7.5% in the activity-deprived cases, and
between 4.4 and 5.9% in the normal development cases. At
160
´ , K. Albus / Developmental Brain Research 123 (2000) 151 – 164
D. Echevarrıa
Fig. 5. Interspike interval (ISI) distributions of spike trains in activity-deprived (recovery) and normally grown (6–14 DIV, 15–51 DIV) OTCs of rat
occipital cortex. The ISI distributions relate to the spike trains shown in Fig. 4A–F (recovery), Fig. 1A–E (6–14 DIV) and Fig. 2A–F (15–51 DIV). The
scale on the left hand side of each histogram gives the number of intervals per bin; binwidth is 10 ms.
6–14 DIV the two oscillating neurones found had a
frequency of 1.1 Hz and 0.7 Hz, respectively.
4. Discussion
4.1. Spontaneous activity and firing rates
The activity types we have defined in OTCs of the rat
occipital cortex have also been recorded by others in
organotypic [11,16,18,19] and explant cultures from rat
neocortex [10,17,20,27] and rat hippocampus [36,55], in
primary dissociated cell cultures from rat neocortex cells
[28,42,44] and rat and mouse hypothalamic neurons
[34,35]. In explant cultures from parietal cortex 3–6 weeks
in vitro the proportion of cells discharging spontaneous
action potentials (20%) was much lower than the proportions (.80%) recorded in our OTCs at that developmental stage. The difference might be due to differences in
the methods for preparation and maintenance of OTCs;
pieces of tissue were 1 mm thick and culturing and
recording was performed under equilibration with 5% CO 2
[54].
The developmental increase in the proportion of spontaneously active neurons we have seen in the neocortical
OTCs is a common phenomenon found in vivo [4,30] and
in vitro [18,20,34,35,42]. The slight developmental decrease in the proportion of bursting neurons reported in
primary dissociated cell cultures of neocortex [28] has not
been confirmed by others [44].
The moderate increase in mean firing rates with time in
vitro is in agreement with another report [11]; according to
our data the increase is not significant, however. Similar
findings were reported from primary dissociated cell
cultures of neocortex [42,44] and hypothalamus [35]. The
firing rates of spontaneously active neurons in our OTCs
are similar to firing rates reported from the isolated
neocortex of the turtle [8] and neocortical primary dissociated cell cultures [42,44]. SBA seems to develop to
higher levels in vivo than in vitro. In the occipital cortex of
´ , K. Albus / Developmental Brain Research 123 (2000) 151 – 164
D. Echevarrıa
161
Fig. 6. Autocorrelograms (ACGs) of spike trains in activity-deprived (recovery) and normally grown (6–14 DIV, 15–51 DIV) OTCs of rat occipital cortex.
The ACGs relate to the spike trains shown in Fig. 4A–F (recovery), Fig. 1A–E (6–14 DIV) and Fig. 2A–F (15–51 DIV). The scale on the left hand side
of each histogram gives the number of action potentials in one bin; binwidth is 10 ms. The two bars at the end of each ACG give the expected value for a
homogeneous distribution, and the chance level (P value50.001), respectively.
the unanesthetized kitten firing rates reach up to 3 Hz
during the second postnatal week and about 8 Hz in the
adult. In the freely moving rat firing rates increase to 3 Hz
at postnatal days 11 and 12 [37] and up to 10 Hz in the
adult [3].
The delay in the development of SBA during the second
week in vitro could be caused by an imbalance in the
growth and differentiation of inhibitory and excitatory
systems and / or cell death of particular neuronal populations [9,32,41]. Habets et al. [28] have presented evidence
that the developmental increase in firing variability in
primary dissociated cell cultures might be due to an arrest
in the maturation of inhibitory interneurones. However
firing variability remained constant during development in
vitro in our OTCs.
We have found that with time in vitro neurons tend to
organize their discharges in the form of discrete bursts. In
addition, the proportions of SpB-neurones as well as the
intensity of the bursts, i.e. the intraburst frequency, increase. The developmental patterns of SBA in OTCs of
neocortex are therefore, similar to that in OTCs of
hippocampus where a progressive time-dependent onset of
spontaneous long lasting group discharges was found
resembling interictal activity in vivo [36,55]. Hippocampal
OTCs older than DIV 30 contained high proportions (60–
100%) of neurons with seizure-like activity.
Seizure-like activity is suggested to develop due to
alterations in synaptic efficiency, neuronal conductances
and / or changes in intrinsic properties of neurons [36].
Interestingly, blockade of inhibition does not seem to be a
prerequisite of seizure-like activity [27]. The spontaneous
synchronous discharges in neocortical OTCs seem to
162
´ , K. Albus / Developmental Brain Research 123 (2000) 151 – 164
D. Echevarrıa
reflect the combined efforts of many neurons distributed all
over the explant; different neurons — although not necessarily connected synaptically — participate in the population phenomenon underlying depolarization shifts [16,27].
4.2. Spontaneous activity as an epigenetic factor in the
development of bioelectric activity in vitro
We have shown that the main effect of activity blockade
is a halt or delay in the development of SBA. Interestingly,
in primary dissociated cell cultures of neocortex [42]
activity blockade increases mean firing rates, intraburst
firing frequencies and burst rates, and decreases burst
duration. Thus, in these cultures the predominant effect of
an activity blockade during development in vitro is a
facilitation of SBA and not a depression as in OTCs of rat
occipital cortex. The difference most probably reflects the
differences in normal development of SBA between both
preparations. In primary dissociated cell cultures the burst
ratio and interval dependencies decrease during development in vitro [28,42]. Spontaneous action potentials mostly
in the form of discrete bursts but at rather long interspike
intervals are observed towards the end of the first week in
vitro and stereotyped burst patterns are present only during
the second week in vitro [28]. In our OTCs higher burst
ratios and stereotyped bursts became dominant only at later
developmental stages. Also, the analysis of autocorrelograms indicated an increase rather than a decrease in the
occurrence of interval dependencies during development in
vitro.
Since the discharge patterns in activity-deprived dissociated cell cultures closely resemble firing patterns seen
after acute administration of picrotoxin it was suggested
that GABA-ergic synaptic inhibition is ineffective in
cortical networks grown under conditions which prevent
SBA [14,42,45]. The finding in OTCs of rat neocortex that
activity blockade strongly reduces the expression of GAD
mRNA supports this idea [46]. Surprisingly, the impairment of the GABA-system did not result in an increase in
firing intensity; in contrast the authors reported significantly reduced average firing rates shortly after the blockade of SBA had been terminated, a finding which is in
accordance with our results.
The halt or delay in the development of SBA in activitydeprived OTCs may be caused therefore, by a retardation
of the structural and functional development of excitatory
systems. This suggestion is supported by anatomical
studies. A significant reduction in the number of dendritic
spines and thus, excitatory synaptic connections in pyramidal neurons has been reported in neocortical OTCs silenced by applying both PTX and TTX [1]. The neuronal
outgrowth and the synapse formation of synaptic ultrastructure is retarded in dissociated cell cultures from
neocortex treated with TTX [52]; retardation of axonal
outgrowth was induced by TTX in the goldfish optic nerve
[23]. As a consequence of such changes the increase in
network density during development in vitro would be
delayed.
The lack of longer lasting tonic burst discharges in most
activity-deprived neurons which is present during the early
recovery period indicates a decrease in the number and
efficiency of both, excitatory inputs and a synaptic NMDA
receptor-mediated component. In normally developing
OTCs this component has been shown to contribute to the
maintenance and amplification of seizure-like events
[27,36,48].
A possible reason for the depression of neuronal activity
induced by chronic exposure to magnesium (Mg 21 ) might
be cell death affecting both excitatory and inhibitory
neurons [7]. However, according to the findings of others
[41] high levels of Mg 21 prevent spontaneous neuronal
degeneration in the hippocampus; widespread neuronal
degeneration occurs only after removal of the blocking
agent. One would expect that a substantial loss of neurons
would reduce the number of neurons detected when
randomly probing the explant with a microelectrode [6].
However, in our activity-deprived OTCs the density of
viable neurons as judged by the proportions of active sites
was around 85% and the same as in normally grown
controls. In addition, the proportions of SpB-neurones and
Sp-neurones had returned to normal values after 5 to 6
days of recovery. The activity depression seen in our
activity-deprived OTCs is therefore, unlikely to be caused
by cell death.
Acknowledgements
We wish to thank Susanne Lausmann for technical
assistance and Erik Aarnoutse and Dr. Gras for providing
software for data evaluation. This work was supported by
the Alexander von Humboldt Foundation (grant
SPK0025), the Gobierno Vasco, Spain (grant no.
BIF94.046) and the BMBF (FKZ 0310954).
References
[1] C.M. Annis, D.K. O’Dowd, R.T. Robertson, Activity-dependent
regulation of dendritic spine density on cortical pyramid neurons in
organotypic slice cultures, J. Neurobiol. 25 (1994) 1483–1493.
[2] A. Antonini, M.P. Stryker, Development of individual geniculocortical arbors in cat striate cortex and effects of binocular impulse
blockade, J. Neurosci. 13 (1993) 3549–3573.
[3] M. Armstrong-James, The functional status and columnar organization of single cells responding to cutaneous stimulation in neonatal
rat somatosensory cortex S1, J. Physiol. 246 (1975) 501–538.
[4] M. Armstrong-James, K. Fox, The physiology of developing cortical
neurons, in: E.G. Jones, A. Peters (Eds.), Development and Maturation of Cerebral Cortex. Cerebral Cortex, Vol. 7, Plenum Press, New
York, 1988, pp. 237–266.
[5] R.E. Baker, M.A. Corner, A.M.M. Habets, Effects of chronic
suppression of bioelectric activity on the development of sensory
´ , K. Albus / Developmental Brain Research 123 (2000) 151 – 164
D. Echevarrıa
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
ganglion evoked responses in spinal cord explants, J. Neurosci. 5
(1984) 1187–1192.
R.E. Baker, M.A. Corner, T. Lammertse, E. Furth, Some functional
effects of suppressing bioelectric activity in fetal mouse spinal
cord-dorsal root ganglion explants, Exp. Neurol. 94 (1986) 426–
430.
R.E. Baker, J.M. Ruijter, Chronic blockade of bioelectric activity in
neonatal rat neocortex in vitro: physiological effects, Int. J. Dev.
Neurosci. 9 (1991) 321–329.
M.G. Blanton, A.R. Kriegstein, Spontaneous action potential activity
and synaptic currents in the embryonic turtle cerebral cortex, J.
Neurosci. 11 (1991) 3907–3923.
M. Caesar, T. Bonhoeffer, J. Bolz, Cellular organization and
development of slice cultures from rat visual cortex, Exp. Brain Res.
77 (1989) 34–244.
M.-C. Calvet, Patterns of spontaneous electrical activity in tissue
cultures of mammalian cerebral cortex vs. cerebellum, Brain Res. 69
(1974) 281–295.
S. Cardoso de Oliveira, A. Reissner, O. Klostermann, P. Wahle,
Development of spontaneous activity in organotypic cultures, in:
¨
22nd Gottingen
Neurobiology Meeting, Vol. 2, 1994, p. 537.
S.M. Catalano, C.J. Shatz, Activity dependent cortical target selection by thalamic axons, Science 281 (1998) 559–562.
M.A. Corner, S.M. Crain, Patterns of spontaneous bioelectric
activity during maturation in culture of fetal rodent medulla and
spinal cord tissues, J. Neurobiol. 3 (1972) 25–45.
M.A. Corner, G.J.A. Ramakers, Spontaneous firing as an epigenetic
factor in brain development. Physiological consequences of chronic
TTX and PTX exposure on cultured rat neocortex neurons, Dev.
Brain Res. 65 (1992) 57–64.
I. Costero, C.M. Pomerat, Cultivation of neurons from the adult
human cerebral and cerebellar cortex, Am. J. Anat. 89 (1951)
405–467.
S.M. Crain, Development of ‘organotypic’ bioelectric activities in
central nervous tissues during maturation in culture, Int. Rev.
Neurobiol. 9 (1966) 1–43.
S.M. Crain, M.B. Bornstein, E.R. Peterson, Maturation of cultured
embryonic CNS tissues during chronic exposure to agents which
prevent bioelectric activity, Brain Res. 8 (1968) 363–372.
S.M. Crain, M.B. Bornstein, Organotypic bioelectric activity in
cultured reaggregates of dissociated rodent brain cells, Science 176
(1972) 182–184.
S.M. Crain, M.B. Bornstein, Early onset of imhibitory functions
during synaptogenesis in fetal mouse brain cultures, Brain Res. 68
(1974) 351–357.
S.M. Crain, Neurophysiologic Studies in Tissue Cultures, Raven
Press, New York, 1976.
J. Del Castillo, L. Engbaek, The nature of the neuromuscular block
produced by magnesium, J. Physiol. 124 (1954) 370–384.
´ K. Albus, Development of spontaneous bioelectric
D. Echevarrıa,
activity in organotypic cultures of rat occipital cortex with and
without spontaneously occurring action potentials, Soc. Neurosci.
Abs. 24 (2) (1998) 1306.
D.L. Edwards, B. Grafstein, Intraocular tetrodotoxin in goldfish
hinders optic nerve generation, Brain Res. 269 (1983) 1–14.
¨
B.H. Gahwiler,
A.M. Mamoon, C.A. Tobias, Spontaneous bioelectric activity in cultured cerebellar purkinje cells during exposure to
agents which prevent synaptic transmission, Brain Res. 53 (1973)
79.
¨
B.H. Gahwiler,
Organotypic monolayer cultures of nervous tissue, J.
Neurosci. Methods 4 (1981) 329–342.
Q. Gu, Y. Liu, R.H. Dyck, V. Booth, M.S. Cynader, Effects of
tetrodotoxin treatment in LGN on neuromodulatory receptor expression in developing visual cortex, Dev. Brain Res. 106 (1998) 93–99.
M.J. Gutnick, B. Wolfson, F.J. Baldino, Synchronized neuronal
activities in neocortical explants cultures, Exp. Brain Res. 76 (1989)
131–140.
163
[28] A.M.M.C. Habets, A.M.J. Van Dongen, F. Van Huizen, M.A. Corner,
Spontaneous neuronal firing patterns in fetal cortical networks
during development in vitro: a quantitative analysis, Exp. Brain Res.
69 (1987) 43–52.
[29] M.J. Hogue, Human fetal brain cells in tissue culture: their
identification and motility, J. Exp. Zool. 106 (1947) 85–107.
[30] P.R. Huttenlocher, Development of cortical neuronal activity in
neonatal cat, Exp. Neurol. 17 (1967) 247–262.
[31] R.R. Kalil, M.W. Dubin, G. Scott, L.A. Stark, Elimination of action
potentials blocks the structural development of retinogeniculate
synapses, Nature 323 (1986) 156–158.
[32] R. Kapoor, C.B. Jaeger, R. Llinas, Electrophysiology of the mammalian cerebellar cortex in organ culture, Neuroscience 26 (1988)
493–507.
[33] T. Kasamatsu, R. Adey, Immediate effects of total visual deafferentation on single unit activity in the visual cortex of freely
behaving cats. I. Tonic excitability changes, Exp. Brain Res. 20
(1974) 157–170.
[34] P. Legendre, A. Tixier-Vidal, J.L. Brigant, J.D. Vicent, Electrophysiology and ultrastructure of mouse hypothalamic neurons in
culture: a correlative analysis during development, Dev. Brain Res.
43 (1988) 273–285.
[35] D.S.F. Ling, R.E. Petroski, W. Chou, M. Geller, Development of
spontaneous electrical activity by rat hypothalamic neurons in
dissociated culture, Dev. Brain Res. 53 (1990) 276–282.
[36] C.J. McBain, P. Boden, R.G. Hill, The rat hippocampal slices in
vitro display spontaneous epileptiform activity following long-term
organotypic cultures, J. Neurosci. Methods 27 (1989) 35–49.
[37] M. Mirmiran, M. Corner, Neuronal discharge patterns in the
occipital cortex of developing rats during active and quiet sleep,
Dev. Brain Res. 3 (1982) 37–48.
[38] T.H. Murphy, L.A. Blatter, W.G. Wier, J.M. Baraban, Spontaneous
synchronous synaptic calcium transients in cultured cortical neurons,
J. Neurosci. 12 (1992) 4834–4845.
¨
[39] G. Paxinos, I. Tork,
L.H. Tecott, K.L. Valentino, Atlas of the
Developing Rat Brain, Academic Press, San Diego, 1991.
[40] A.A. Penn, P.A. Riquelme, M.B. Feller, C.J. Shatz, Competition in
retinogeniculate patterning driven by spontaneous activity, Science
279 (1998) 2108–2112.
[41] L.D. Pozzo Miller, N.K. Mahanty, J.A. Connor, D.M. Landis,
Spontaneous pyramidal cell death in organotypic slice cultures from
rat hippocampus is prevented by glutamate receptor antagonists,
Neuroscience 63 (1994) 471–487.
[42] G.J.A. Ramakers, M.A. Corner, A.M.M.C. Habets, Development in
the absence of spontaneous bioelectric activity results in increased
stereotyped burst firing in cultures of dissociated cerebral cortex,
Exp. Brain Res. 79 (1990) 157–166.
[43] G.J.A. Ramakers, G.J. Boer, Chronic suppression of bioelectric
activity and cell survival in primary cultures of rat cerebral cortex:
biochemical observations, Eur. J. Neurosci. 3 (1991) 154–161.
[44] G.J.A. Ramakers, C. de Wit, P.S. Wolters, M.A. Corner, A developmental decrease in NMDA-mediated spontaneous firing in
cultured rat cerebral cortex, Int. J. Dev. Neurosci. 11 (1993) 25–32.
[45] G.J.A. Ramakers, H. Van Galen, M.G. Feenstra, M.A. Corner, G.J.
Boer, Activity-dependent plasticity of inhibitory and excitatory
amino acid transmitter systems in cultured rat cerebral cortex, Int. J.
Dev. Neurosci. 12 (1994) 611–621.
[46] A. Reissner, S. Cardoso de Oliveira, O. Klostermann, P. Wahle,
Blockade of spontaneous activity in rat occipital cortex cultures
¨
downregulates GAD mRNA expression, in: 23rd Gottingen
Neurobiology Meeting, Vol. 2, 1995, p. 102.
[47] R.V. Riccio, M.A. Matthews, The postnatal development of the rat
primary visual cortex during optic nerve impulse blockade by
intraocular tetrodotoxin: a quantitative electron microscopic analysis, Dev. Brain Res. 20 (1985) 55–68.
[48] M.M. Segal, Epileptiform activity in microcultures containing one
excitatory hippocampal neuron, J. Neurophysiol. 65 (1991) 761–
770.
164
´ , K. Albus / Developmental Brain Research 123 (2000) 151 – 164
D. Echevarrıa
[49] F.J. Seil, R. Drake-Baumann, Reduced cortical inhibitory synaptogenesis in organotypic cerebellar cultures developing in the
absence of neuronal activity, J. Comp. Neurol. 342 (1994) 366–377.
[50] M.P. Stryker, W.A. Harris, Binocular impulse blockade prevents the
formation of ocular dominance columns in cat visual cortex, J.
Neurosci. 6 (1986) 2117–2133.
[51] I. Thompson, C. Holt, Effects of intraocular tetrodotoxin on the
development of the retinocollicular pathway in the Syrian hamster,
J. Comp. Neurol. 282 (1989) 371–388.
[52] F. Van Huizen, H.J. Romijn, A.M.M.C. Habets, Synaptogenesis in
rat cerebral cortex is affected during chronic blockade of spontaneous bioelectric activity by TTX, Dev. Brain Res. 19 (1985)
67–80.
[53] J.M. Woijtowicz, K.C. Marshall, W.J. Hendelman, Depression by
magnesium ion of neuronal excitability in tissue culture of central
nervous system, Can. J. Physiol. Pharmacol. 55 (1977) 367–372.
[54] B. Wolfson, M.J. Gutnick, F.J. Baldino, Electrophysiological characteristics of neurons in neocortical cultures, Exp. Brain Res. 76
(1989) 122–130.
[55] B. Zipser, S.M. Crain, M.B. Bornstein, Directly evoked paroxymal
depolarizations of mouse hippocampal neurons in synaptically
organized explants in longterm cultures, Brain Res. 60 (1973)
489–495.