Skip to main content
PMC home page
The Journal of Physiology logoLink to The Journal of Physiology
. 2004 Aug 5;560(Pt 2):377–390. doi: 10.1113/jphysiol.2004.071621

Spontaneous, synchronous electrical activity in neonatal mouse cortical neurones

Rebekah Corlew 1, Martha M Bosma 1, William J Moody 1
PMCID: PMC1665264  PMID: 15297578

Abstract

Spontaneous [Ca2+]i transients were measured in the mouse neocortex from embryonic day 16 (E16) to postnatal day 6 (P6). On the day of birth (P0), cortical neurones generated widespread, highly synchronous [Ca2+]i transients over large areas. On average, 52% of neurones participated in these transients, and in 20% of slices, an average of 80% participated. These transients were blocked by TTX and nifedipine, indicating that they resulted from Ca2+ influx during electrical activity, and occurred at a mean frequency of 0.91 min−1. The occurrence of this activity was highly centred at P0: at E16 and P2 an average of only 15% and 24% of neurones, respectively, participated in synchronous transients, and they occurred at much lower frequencies at both E16 and P2 than at P0. The overall frequency of [Ca2+]i transients in individual cells did not change between E16 and P2, just the degree of their synchronicity. The onset of this spontaneous, synchronous activity correlated with a large increase in Na+ current density that occurred just before P0, and its cessation with a large decrease in resting resistance that occurred just after P2. This widespread, synchronous activity may serve a variety of functions in the neonatal nervous system.

Spontaneous electrical activity regulates many of the most fundamental processes of nervous system development, including neuronal migration (Komuro & Rakic, 1992, 1998), axonal outgrowth (Gu et al. 1994; Catalano & Shatz, 1998), transmitter phenotype selection (Gu et al. 1994), dendritic patterning (Wong & Ghosh, 2002), activation of transmitter receptors (Liao et al. 2001), synaptic pruning (Shatz & Stryker, 1988; O'Leary et al. 1994), expression of, release of, and receptivity to neurotrophins (Blochl & Thoenen, 1995; Meyer-Franke et al. 1998; Shieh & Ghosh, 1999), programmed cell death (Svoboda et al. 2001), and development of mature ion channel types (Desarmenien & Spitzer, 1991; Linsdell & Moody, 1994, 1995; Dallman et al. 1998; Grosse et al. 2000). The first step in the transduction of spontaneous electrical activity into its developmental sequelae is the transient increase in intracellular Ca2+ concentration ([Ca2+]i), caused by the influx of Ca2+ through either voltage-dependent Ca2+ channels or NMDA receptors (Spitzer, 2002).

One of the most common forms of spontaneous activity early in brain development is the synchronous occurrence of [Ca2+]i transients and bursts of action potentials among large neuronal populations. This type of widespread synchronized activity occurs in the retina (Meister et al. 1991), hippocampus (Garaschuk et al. 1998), spinal cord (Wenner & O'Donovan, 2001), and in the neocortex (Garaschuk et al. 2000; Sisk & Moody, 2003). These bursts occur at relatively low frequencies, ranging from about 1 burst min−1 in hippocampus, to about 1 burst (10 min)−1 in spinal cord (O'Donovan, 1999). Although the significance of this widespread synchronicity is not well understood in most cell types, in retina it serves, among other functions, to encode the identity and location of retinal ganglion cells so that their appropriate synaptic connections are formed in the lateral geniculate nucleus (Wong, 1999).

Spontaneous activity in developing neurones and muscle cells can occur in a completely cell-autonomous manner, in isolated cells, regulated by the intrinsic patterns of ion channel development (Greaves et al. 1996). In complex neuronal circuits developing in vivo, however, such activity is likely to be caused, as well as synchronized, by a combination of intrinsic electrical properties of neurones and the emerging electrical and chemical synaptic circuitry that connects them (Gust et al. 2003). The exact patterns with which various ion channel types are regulated in developing cells can be critical in determining both the occurrence and the properties of spontaneous activity. In developing ascidian muscle, for example, the temporary downregulation of the inwardly rectifying K+ current early in development triggers spontaneous action potentials whose bursting properties and ability to mediate waves of Ca2+ influx are determined by the Ca2+ and outward K+ currents that appear at the same stage (Greaves et al. 1996; Dallman et al. 1998, 2000). Relatively little is known about the role of voltage-gated ion channels in regulating spontaneous activity during CNS development, in part because there are few cell types in which the developmental onset and cessation of spontaneous activity and the development of voltage-gated ion channels have been studied during the same stages.

Here we report that spontaneous and highly synchronous [Ca2+]i transients occur in mouse cortical neurones primarily on the day of birth. These [Ca2+]i transients result from Ca2+ influx during electrical activity. The developmental timing of this widespread, synchronous activity suggests that its onset is regulated by a large increase in Na+ current density and its cessation by a large decrease in resting resistance that occurs a few days later in development.

A preliminary report of these findings has been published previously (Sisk & Moody, 2003).

Methods

Animals and tissue preparation, solutions, drugs

Time-mated Swiss-Webster mice were purchased from ATL-Harlan (Kent, WA, USA) and maintained in the laboratory until the appropriate stage. For embryonic stages, pregnant mice were killed by exposure to a rising concentration of CO2, fetuses removed and brains excised quickly and placed in ice-cold artificial cerebrospinal fluid (ACSF; mm: 119 NaCl, 5 KCl, 1.3 MgCl2, 2.5 CaCl2, 1 NaHPO4, 26.2 NaHCO3, 11 d-glucose) bubbled with 95% O2–5% CO2. For postnatal stages, pups were anaesthetized by hypothermia and killed by decapitation, and brains removed as above. All procedures were approved by the University of Washington Institutional Animal Care and Use Committee and follow US federal guidelines. Horizontal slices, 400 μm thick, were cut in ice-cold ACSF using a Vibratome (Technical Products International, St Louis, MO, USA) and were allowed to recover at 35°C for 1 h in oxygenated ACSF.

Tetrodotoxin (TTX), tetraethylammonium (TEA), 4-aminopyridine (4-AP), and nifedipine were all obtained from Sigma-Aldrich (St Louis, MO, USA). TTX and TEA were added directly to the ACSF. Nifedipine was diluted into ACSF from a 10 mm stock solution in DMSO.

Ca2+ imaging

After recovery, slices were incubated for 15 min at room temperature in oxygenated ACSF containing the [Ca2+]i-indicating dye fluo-4 (1.75 μm) and 0.07% Pluronic F-127 (Molecular Probes, Eugene, OR, USA). Slices were rinsed and then placed into a glass-bottomed chamber on the stage of an inverted microscope (Nikon Diaphot) equipped with a × 20 fluorescence objective, and were continuously superfused with oxygenated ACSF at approximately 1 ml min−1 at 30–33°C. Temperature was maintained by a combination of an in-line heater on the solution inflow tubing and a temperature-controlled chamber holder (Warner Instruments, Hamden, CT, USA). Images were captured with a cooled CCD camera (Photometrics, Tucson, AZ, USA) using a capture time of 300 ms at an interval of 0.5–1.0 s, using MetaFluor software (Universal Imaging, West Chester, PA, USA).

At the start of each experiment, an area of the cortex was selected for imaging. Areas were generally in the lateral aspect of the horizontal slice, always lateral to a line connecting the frontal pole with the most posterior part of the entorhinal cortex (see diagram Fig. 1A), with some bias toward areas midway between the frontal and occipital poles. The box in Fig. 1A indicates a typical area chosen for imaging, which was rectangular, 700 μm wide and 892 μm deep. For each experiment, 27–116 fluo-4-filled individual cell bodies were selected, spanning the complete depth and width of the imaged area. Each cell (‘region’) was marked with a circular or ovoid surround, and the software captured the average fluorescence within that region at each time point. Thus each individual trace in the records shown represents the fluo-4 fluorescence within a single cell as a function of time. In each experiment, several off-tissue regions of the same size as the regions marking cells were chosen to serve as controls. Figure 1B shows a camera image from a typical experiment.

Figure 1. Horizontal slice preparation.

Figure 1

Open in a new tab

A, diagram of a horizontal slice taken from a P0 mouse brain; A, anterior; P, posterior. The arrows indicate the boundaries within which the imaged area was chosen, and the box indicates a typical area chosen for imaging. B, camera image of a fluo-4-loaded slice area used in the imaging experiments. Scale bar, 100 μm.

Data analysis

For analysis, all traces from each experiment were exported in text format into SigmaPlot (SPSS, Chicago, IL, USA). Each trace (fluorescence records from individual cell bodies) was first smoothed using a software high-pass filter to remove baseline drift. Filtered traces were displayed and examined individually for a general impression of overall level of activity and synchrony, and the amplitudes, duration, and frequency of the activity was measured. For quantitative analysis of patterns of activity in individual cells and of synchrony among cells, traces were idealized using a custom-written threshold detection routine based on a ΔF/F event detection. Each point in the data that fell above the detection criterion was converted to 1.0 and each point below threshold to 0.0. This produced a binary event trace for each cell of each experiment (see Gust et al. 2003). These idealized records were used to count the transients in each individual cell and calculate an average frequency both for that cell and for all cells in an experiment.

To quantify synchrony, all idealized traces for a given experiment were averaged, producing a summary record in which each transient had an amplitude equal to the fraction of cells that participated in that [Ca2+]i transient. So, for example, if 10% of the cells imaged participated synchronously in a given transient, that transient appears in the averaged idealized record with an amplitude of 0.1. Figure 2D shows the idealized summary record for a part of the experiment shown in Fig. 2C.

Figure 2. Spontaneous, synchronous [Ca2+]i transients in the P0 mouse cortex.

Figure 2

Open in a new tab

A and B, spontaneous [Ca2+]i transients from a slice in which 48 neurones were imaged for 110 min. A shows data from six cells for 55 min of the recording. At 86 min (bar), 50 mm KCl was applied to the slice. B shows six different cells from the same slice at a higher time resolution to emphasize the high degree of synchronicity. C, six neurones from another P0 slice in which activity was less well resolved, but still clearly visible. D, idealized summary record including all 99 cells imaged, displayed on the same time axis as C. See Methods for procedures for idealization of records and creation of the summary record. In A–C (and in all subsequent records in this paper), the baseline fluorescence for some cells was adjusted slightly to offset the traces for clarity.

To characterize synchronous activity in a given slice, we extracted the following three parameters from this summary record.

  1. The mean fraction of cells that participated in synchronous activity. We term this the synchronicity index (SI) for the experiment. It is derived by excluding all non-synchronous transients (< 20% of cells participating in that transient) and then calculating the mean fraction of cells participating in all remaining transients.

  2. The product of the SI and mean frequency of all transients with ≥ 20% participation (SI × Freq). Some slices generate only a few transients, but in these many cells participate, yielding high SI values. The SI × Freq parameter weights the SI values by the overall intensity of the activity, and the combination of SI and SI × Freq gives a good overall indication of the amount of synchronous activity in a given preparation.

  3. The mean frequency of transients with amplitudes greater than 0.8 (80% of cells participating). Activity with this high degree of synchronicity was rare at stages other than P0, and this parameter allowed us to quantify that developmental time course (see below).

For the analysis of the frequency of occurrence of [Ca2+]i transients independent of whether they were synchronous among cells, we summed all transients from all individual idealized records in an experiment, not just those that were synchronous, and computed overall transient frequency as transients per cell per minute. This analysis was used, for example, in Figs 6D, and 8C and D, to ask whether the developmental appearance of synchronous [Ca2+]i transients reflected the development of the transients themselves or of synchronization among already existing transients.

Figure 6. Development of parameters of spontaneous activity between E16 and P6.

Figure 6

Open in a new tab

A, synchronicity index (mean fraction of cells participating in synchronous activity; see text). B, synchronicity index times the mean frequency of transients with greater than 20% of cells participating (see text for details). C, mean frequency of transients in which > 80% of cells participated. In A–C, data from stages later than P2 were combined. Single and double asterisks indicate significance at P values of ≤ 0.05 and ≤ 0.01, respectively. n values are: E16 (n = 6), E17 (n = 4), E18 (n = 7), P0 (n = 60), P1 (n = 17), P2 (n = 7), > P2 (n = 6). D, mean frequency of small transients in individual cells, independent of whether they occurred synchronously with those in other cells. See Methods. Data from stages earlier than P0 and later than P0 were combined for this plot. The SI values for the subset of cells analysed for this plot were: < P0, 0.21 ± 0.05; P0, 0.65 ± 0.05; and > P0, 0.30 ± 0.05. The SI values at < P0 and > P0 were each significantly different from that at P0: P < 0.001. This indicates that low synchronicity values seen at stages other than P0 are not due simply to lack of activity.

Figure 8. Large asynchronous and small synchronous [Ca2+]i transients.

Figure 8

Open in a new tab

A, coexistence of small synchronous and large asynchronous [Ca2+]i transients in individual cells in a P0 slice. Note high degree of synchronicity and the higher frequency of small transients, which was typical. B, block of small, but not large transients by TTX in an E16 slice. Note that as was typical of E16 slices, small transients were not synchronous. C, changes in the occurrence of large transients with development. For this analysis, large transients were counted as they were for Fig. 6D, i.e. in individual cells whether or not they were synchronous with other cells. See Methods for detection procedure. D, change in the ratio of small/large transients with development. For this plot, small and large transients were counted in individual cells whether or not they were synchronous, and the ratio of the two computed separately for each cell and then averaged. n values were: < P0, 11; P0, 20; > P0, 23. Single and double asterisks indicate P ≤ 0.05 and P ≤ 0.01, respectively. NS, not significant.

In the analyses of small synchronous [Ca2+]i transients versus large asynchronous transients (see Figs 6D, and 8C and D), large and small transients were separated using a window discriminating routine that separated ‘small’ transients (amplitude greater than ΔF/F = 0.05 and less than ΔF/F = 0.1) from ‘large’ transients (amplitude greater than ΔF/F = 0.1). After separation, the two types of transients were analysed as above.

Extracellular recording

Micropipettes were pulled to resistances from 0.2 to 0.5 MΩ (measured in ACSF) and filled with ACSF. The extracellular signal was amplified with a AC-coupled amplifier (AM Systems, Everett, WA, USA) and band-pass filtered at 0.1 Hz to 1 kHz. The amplified signal was digitized at 2 kHz using pCLAMP 8.0 software (Axon Instruments, Foster City, CA, USA).

Statistical significance was calculated using Student's t test, either a paired test (when single slices were exposed to different conditions) or an unpaired heteroscedastic test for unpaired data, or using Fisher's exact test (2-tailed) for the clustering data. Data are presented as mean ± s.e.m.

Results

Spontaneous, synchronous [Ca2+]i transients in the P0 mouse cortex

We conducted successful imaging experiments on 60 neocortical slices taken from animals in 27 litters at P0, the day of birth. In each slice, data were collected from 27 to 116 cells (mean, 67), plus off-tissue blanks as controls to confirm that transients were not artifactual.

The majority of P0 slices showed spontaneous [Ca2+]i transients that were highly synchronized across a large percentage of neurones imaged in a single 20× field of view (area imaged: 700 μm × 892 μm). Figure 2A shows an example of such activity recorded from a slice that showed a particularly high degree of synchronization of the spontaneous activity. In this slice, 48 cells were imaged for 110 min (only 45–100 min are shown). Spontaneous [Ca2+]i transients occurred throughout this period at a mean frequency of 0.6 min−1. Off-tissue blank regions showed no transients. Between 95 and 100% of imaged cells participated in these transients. To test whether these [Ca2+]i transients were likely to result from electrical activity and Ca2+ entry, 50 mm KCl was applied to the extracellular solution. As expected if the [Ca2+]i transients resulted from action potential activity, this resulted in a brief burst of transients in most cells superimposed on a large increase in baseline [Ca2+]i, followed by a cessation of activity.

Figure 2B shows several of the transients from six different cells within the same preparation at higher time resolution, emphasizing the high degree of synchrony. With our image capture interval of 300 ms we were unable to detect any difference in the time of onset of the transients among different cells. Spontaneous transients ranged in duration from 5 to 15 s, and in amplitude from 0.25 to 18% ΔF/F. In highly synchronous slices such as that shown in Fig. 2A synchronous activity represented a large fraction of the total activity generated by the cells. In the slice shown in Fig. 2A, for example, during a 30 min subset of the recording time just before KCl application, the 48 cells generated a total of 1182 individual [Ca2+]i transients above our detection threshold. Of these 863 were synchronous in ≥ 90% of the cells. (The remaining activity was a combination of similar transients in which 20–89% of the cells participated, smaller transients in which only one or two cells participated (visible in Fig. 2B), and very large transients in single cells. See below.) Figure 2C shows another example of this widespread, synchronous activity taken from a different animal, also at P0. Figure 2D shows the idealized summary record for a part of the experiment shown in Fig. 2C, created using the analysis procedures described in Methods.

At P0, 75% (45/60) of slices showed spontaneous, synchronous activity, defined as the occurrence at a frequency of at least 0.2 min−1 of transients in which at least 20% of the cells participated. In those 45 slices, the mean SI was 0.58 ± 0.03, so that on average 58% of the neurones participated in synchronous activity (range 20–100%). The mean frequency of those transients was 0.91 ± 0.12 min−1.

To confirm that these spontaneous [Ca2+]i transients were caused by electrical activity, we applied tetrodotoxin (TTX; 1 μm) to six slices, and in each case it reversibly blocked the transients (Fig. 3A). In these slices, TTX reduced the SI from 0.67 ± 0.06 to 0.13 ± 0.06 (P = 0.0001), and also reduced the frequency of these transients, so that the SI × Freq parameter was reduced from 0.97 ± 0.26 (Control) to 0.02 ± 0.01 (TTX) (P = 0.01). Activity was also blocked by 500 μm nifedipine (Fig. 3B; n = 3), indicating that at least some of the Ca2+ influx occurs via voltage-gated, probably L-type, Ca2+ channels.

Figure 3. Blocker sensitivity of synchronous [Ca2+]i transients.

Figure 3

Open in a new tab

A, five cells from a recording in a P0 slice showing that spontaneous [Ca2+]i transients are blocked by TTX (1 μm). B, five cells from a recording in a P0 slice showing block of spontaneous [Ca2+]i transients by nifedipine (500 μm). C, extracellular microelectrode recording of spontaneous electrical activity in a P0 slice, showing spontaneous activity that is blocked by TTX. D, block of spontaneous [Ca2+]i transients by octanol (3 mm).

As further confirmation that the synchronous [Ca2+]i transients resulted from electrical activity, we made extracellular microelectrode recordings. These showed spontaneous activity (Fig. 3C; n = 5 out of 8 slices) consisting of biphasic transients 20–400 μV in amplitude occurring at a mean frequency of 0.6 ± 0.37 Hz, very similar to that of the spontaneous [Ca2+]i transients. This extracellularly recorded activity was blocked by TTX (Fig. 3C; n = 3) and either induced or potentiated by TEA (n = 2; see below).

In addition, the gap junction blockers octanol and heptanol (3 mm) blocked the activity (Fig. 3D; n = 3), suggesting that gap-junctional communication among cortical neurones, or among a subset of pacemaker neurones, is required for synchronous activity. We have previously shown that octanol blocks dye coupling via gap junctions in mouse embryonic ventricular zone cells, but does not affect voltage-gated sodium or potassium currents (Picken-Bahrey & Moody, 2003a), making it more likely that octanol exerts its effects on synchronous activity via its action on gap junctions.

We next asked whether, in slices with low synchronicity indices, groups of synchronously active cells could be identified and if so, whether those cells were spatially clustered. We identified 13 slices in which SI values were less than the mean value (range 0.27–0.52) and in which groups of data traces could be found that were synchronously active with each other. (This represented 50% of the slices with SI values less than 0.52. In the other slices, low SI values resulted from various individual cells that participated in some transients but not in others, or from cells with visually apparent synchronous transients that failed our threshold detection routine; in these slices no synchronously active subgroups of traces could be identified.) In these 13 slices there were 22 groups of cells (mean size: 11.8 ± 2.8 cells out of a mean total number of cells in the slice of 54.5 ± 4.3). The x–y coordinates of each cell in the slice were used to construct a spatial plot of all cells, with synchronously active cells identified. We then constructed by eye an elliptical region to enclose a minimum of 90% of the synchronously active cells (mean 98.4 ± 0.7%). In each of the 22 cases, an elliptical boundary that enclosed the active cells and excluded the majority of inactive cells could be constructed, and in each case that area contained a higher ratio of active/inactive cells than the remainder of the imaged area (2-tailed Fisher's exact test; mean P value 0.002, range 2.0 × 10−13 to 0.002). Figure 4 shows an example of a plot of cell locations indicating a cluster of synchronously active cells in a large field of cells that did not participate in synchronous activity. This indicates that, at least in half the slices in which fewer than 50% of the cells participate in synchronous activity, those cells that do are spatially clustered. In four slices, multiple clusters of cells could be identified (2–5 clusters per slice); cells within a single cluster were synchronously active with each other, but not with cells of other clusters or with the general population of cells. In each of these slices, there was no spatial overlap between the clusters of synchronously active cells. (In slices with > 52% SI values, no non-uniform distribution of synchronously active cells was detected.)

Figure 4. Clusters of synchronously active cells in P0 cortex.

Figure 4

Open in a new tab

Plot of the location of cells in a P0 slice that showed a small cluster of synchronously active cells. The plot approximates the entire area imaged and contains all imaged cells. The pia is indicated by a continuous line. •, cells that participated in synchronous activity; ○, cells that did not. The synchronously active cells are surrounded by the elliptical region used for cluster analysis (see text).

The synchronously active clusters of cells showed no preferential location within the cortical layers. We overlayed each plot of cell clusters generated for the above analysis on the camera image of the experiment and divided the cortex into four bands 136 μm thick, parallel to the pial surface. Summing cells from all of the 13 slices by band, we found the following fraction of synchronously active cells in each band: Band 1 (pial surface), 0.30 (69/158); Band 2, 0.32 (78/246); Band 3, 0.41 (45/111); Band 4, 0.29 (8/28). Pairwise Fisher's exact tests showed no significant differences between these fractions for any two bands (P value range: 0.07–1.0).

Development of spontaneous, synchronous activity

To measure the developmental profile of activity, we repeated the above imaging experiments in slices taken from animals at E16 (n = 6), E17 (n = 4), E18 (n = 7), P0 (n = 60), P1 (n = 17), P2 (n = 7), P4 (n = 1), P5 (n = 3) and P6 (n = 2). Representative records from slices at several of these stages are shown in Fig. 5. Figure 6A–C shows the values of SI, SI × Freq, and the frequency of transients in which > 80% of cells participate, each as a function of developmental stage. These experiments showed that spontaneous, synchronous activity is significantly centred at P0, the day of birth. For example, the mean SI increases from 0.15 ± 0.07 at E16 to 0.52 ± 0.03 at P0 (P = 0.001) and then back to 0.24 ± 0.07 at P2 (P = 0.006) (Fig. 6A). (The SI values at P0 include slices which showed no synchronous activity, and so the value at P0 is slightly lower than that cited above.) The mean frequency of synchronous transients was also highest at P0, so the SI × Freq parameter, the best measurement of the overall intensity of synchronous activity, shows a sharper peak at P0 than does the SI parameter alone. SI × Freq increases from 0.12 ± 0.09 at E16 to 0.59 ± 0.09 at P0 (P = 0.003) and decreases again to 0.04 ± 0.01 by P2 (P < 0.001)(Fig. 6B). The most dramatic example of how activity occurs preferentially at P0 was seen in the frequency of spontaneous transients in which 80% or more of the cells participated. At P0, these occurred at a mean frequency of 0.38 ± 0.08 min−1, whereas they were rarely seen at all at any stage before or after P0 (Fig. 6C).

Figure 5. Development of spontaneous activity.

Figure 5

Open in a new tab

Representative records from slices at E16, E18, P0 and P1 are shown. Note that at E16, substantial activity occurs, even though synchronicity of the activity is low. At P1, two large transients are noted, as discussed in Results.

Records like those in Fig. 5 (E16 and E18) suggest that the lower levels of synchronicity at stages earlier than P0 are due, at least in part, to a lack of synchronization of activity that is already occurring, rather than a general lack of spontaneous activity. To test this hypothesis, we chose a subset of slices representing stages earlier than P0 (n = 11), P0 (n = 20), and later than P0 (n = 23). These slices were chosen as representative of the mean SI values at these groups of stages (SI values of selected slices: < P0, 0.21 ± 0.05; P0, 0.65 ± 0.05; > P0, 0.30 ± 0.05). For each cell in each slice, we counted the total number of transients in each idealized record for each cell, independent of whether they were synchronous with transients in other cells. The total number of transients for all cells of the experiment was then determined, and the mean frequency for that experiment expressed as transients cell−1 min−1. We found no significant difference in the overall frequency of [Ca2+]i transients in individual cells with development (Fig. 6D), indicating that, at least in some cells, spontaneous activity develops before the mechanisms that synchronize it and persists after the mechanisms that synchronize it decline. Even though they are not synchronous, [Ca2+]i transients in individual cells at stages earlier than P0 are TTX sensitive (n = 3; data not shown).

Effect of TEA on synchronous activity at P0

Out of our sample of 60 P0 slices, 4 showed no synchronization at all (SI = 0). To determine whether synchronous activity could be induced in these slices by artificially increasing neuronal excitability, we applied either TEA (10 mm) or 4-AP (1 mm). In each of the four slices, TEA or 4-AP caused the immediate appearance of synchronized activity (Fig. 7A and B). The parameters characterizing this TEA-induced activity (SI = 0.80 ± 0.1; SI × Freq = 0.85 ± 0.15) were not significantly different from those for the total population of slices under control conditions (P = 0.07 and 0.19, respectively). (These parameters were obviously significantly higher than in the paired controls, which were selected to show no synchronous activity: SI = SI × Freq = 0.) TEA or 4-AP were also applied to six other P0 slices that did show modest levels of synchronous activity, and in each case activity was increased in both average synchronicity and in frequency. For the total sample of 10 P0 slices, the effect of TEA on both SI and SI × Freq was highly significant (TEA versus control: SI, 0.72 ± 0.06 versus 0.19 ± 0.06, P < 0.0002; SI × Freq, 0.96 ± 0.15 versus 0.12 ± 0.06, P = 0.0002) (Fig. 7C and D). To confirm that these drugs were indeed eliciting synchronous activity via their effects on neuronal excitability, we confirmed that TTX (1 μm) reversibly blocked the activity that they elicited (Fig. 7A and B).

Figure 7. Effects of TEA on spontaneous [Ca2+]i transients.

Figure 7

Open in a new tab

A, induction of synchronous [Ca2+]i transients in a P0 slice by TEA and reversible block of TEA-induced activity by TTX. This slice represents the group of P0 slices selected to study the effects of TEA because of their low levels of spontaneous [Ca2+]i transients. B, idealized summary record of all 94 cells imaged in this slice, displayed on the same time axis as A. C and D, summary of effects of TEA on SI and SI × Freq in slices at E16 (n = 6), E17 (n = 4) and P0 (n = 10). C, control.

Effect of TEA on synchronous activity at stages earlier than P0

Since TEA at the concentration we used produced synchronous activity in 100% of the P0 slices we studied (10/10), we asked whether TEA exposure could produce synchronous activity at earlier stages, in which spontaneous synchronous activity occurred less frequently. We exposed slices to TEA at E16 (n = 6) and E17 (n = 4), when spontaneous synchronous activity was rare (see above). (These slices were typical in their low levels of synchronous activity, unlike the P0 slices exposed to TEA, which were chosen for their atypical low levels of synchronicity.) The results (Fig. 7C) show that at E16, TEA did not increase SI values (P = 0.52) whereas at E17, the increase caused by TEA is significant (P = 0.03) but much smaller than its effect at P0 (P < 0.0002). A similar pattern was seen with SI × Freq values (Fig. 7D). We did not study the effects of TEA at later stages.

Small and large spontaneous transients

In addition to the small transients that comprised synchronous activity in many preparations, we also observed much larger [Ca2+]i transients in some slices at all stages studied (Fig. 8A). These transients were sometimes brief in duration, much like the small transients, but in other cells were much longer in duration. These large transients occurred much less frequently than the small, synchronous transients, and rarely occurred synchronously even in small subsets of cells.

Although large transients occurred at all stages (see below), only in five slices (1 at P0, 1 at P2, 1 at P4, 2 at P6) were they present at high enough frequencies for meaningful analysis of SI and frequency. The mean SI × Freq for large transients in these five slices was 0.006 ± 0.004, which is significantly lower than the SI × Freq parameter at P0 (0.59 ± 0.09; P < 0.001) or at P2 (its lowest value for small transients: 0.04 ± 0.01; P < 0.001). In two of these slices (1 at P0, 1 at P4), both large and small transients occurred together at a high enough frequency to make these measurements. The SI × Freq parameter for the large transients in these two slices were 0.022 and 0, respectively; the SI × Freq parameter for the small transients in these two slices were 0.64 and 0.16, respectively.

Although both types of transients often occurred in the same slice (or even in the same cell), they had very different developmental profiles. We analysed the occurrence of both types of transients in 54 slices (the same sample used for the analysis shown the Fig. 6D) using a window discriminator software routine (see Methods). The results of this analysis are shown in Fig. 8C and D. As shown above (Fig. 6D), the overall frequency of small transients in individual cells did not change with development. Large transients, however, were more frequent after P0 than at or before P0 (Fig. 8C). When the ratios of small/large transients were computed individually for each preparation and the ratio averaged, the results showed that small transients comprised a much larger fraction of total transients at P0 that at any other stage (Fig. 8D).

The low frequency of occurrence and lack of synchronicity of large transients suggest that they may be generated by a different basic mechanism from the small transients. To test whether the large transients resulted from electrical activity in the form of Na+-dependent action potentials, we exposed 10 slices (6 animals, 6 litters) that showed a relatively high frequency of large transients to TTX. The large transients were not blocked by TTX, including four cases (one of which is shown in Fig. 8B) in which we recorded both large asynchronous and small synchronous transients in the same slice at a high enough frequency to confirm directly that the latter were resistant to TTX block while the former were sensitive.

Correlation of synchronized activity with developmental changes in intrinsic neuronal properties

In our experiments, spontaneous, synchronized activity is much more common at P0 than even one day earlier or later. To see whether the developmental profile of spontaneous synchronized activity is correlated with changes in intrinsic electrophysiological properties, we compared our present data on synchronized activity with our previous data on Na+ current amplitude and input resistance of cortical neurones at these stages (Picken-Bahrey & Moody, 2003b). Figure 9 shows an overlay of synchronicity index, INa amplitude, and input resistance measured from E16 to > P2. INa increases monotonically from E16 to P2, and the onset of spontaneous synchronized activity closely parallels this rise. INa does not decrease as activity disappears after P0, but the input resistance of the neurones decreases dramatically, which would have the effect of reducing neuronal responses to both synaptic inputs and to their own Na+ current. These data suggest the possibility that the window of development during which spontaneous, synchronized activity occurs is determined by those stages at which both INa density and input resistance are high.

Figure 9. Relationship between spontaneous, synchronous activity and the development of Na+ current and input resistance.

Figure 9

Open in a new tab

Synchronicity index (grey columns), Na+ current amplitude (○), and input resistance (Rin,•), as functions of development in cortical neurones. Na+ currents are plotted in arbitrary units and input resistance in GΩ/10. Na+ current and input resistance data are taken from Picken-Bahrey & Moody (2003b). Note that spontaneous, synchronous activity occurs during a window of development in which both INa and Rin are high.

Discussion

Widespread synchronous spontaneous activity in the developing mouse neocortex

We have shown that neurones of the embryonic and neonatal mouse neocortex generate spontaneous [Ca2+]i transients at a mean frequency of about 1 min−1 that are highly synchronous across large populations of cells. At P0, on average, 59% of the neurones participate in this activity, with 80–100% participating in many slices. These spontaneous [Ca2+]i transients are blocked by TTX or nifedipine, indicating that they result from spontaneous electrical activity involving TTX-sensitive Na+ channels leading to Ca2+ influx through voltage-gated, probably L-type, Ca2+ channels.

Developmental onset and cessation of synchronous activity

This widespread, synchronous activity occurs during a narrow range of developmental stages, centred at P0, the day of birth. The appearance of widespread, synchronous activity between E16 and P0 was reflected in each of the parameters characterizing that activity that we measured. Thus, between E16 and P0 the average fraction of cells participating in activity rose almost 4-fold, from 15% to over 50%, and the mean frequency of synchronous transients more than doubled, from 0.4 min−1 to 0.9 min−1. At E16, we detected no transients in which 80% or more of the cells participated, whereas at P0 these were common, occurring at a mean frequency of 0.4 min−1.

The appearance of synchronous activity seems to be due more to the development of mechanisms providing synchronicity than to the appearance of spontaneous [Ca2+]i transients themselves, which occur at about the same overall rate in individual cells at E16 as at P0 (see Fig. 6D). Our data also suggest that the onset of synchronous activity may be caused in part by the increase in Na+ current density that occurs over this time period (Fig. 9). Thus the increase in Na+ current may not simply be serving to render the cells excitable enough to generate spontaneous [Ca2+]i transients, but rather serving a role in synchronizing activity among neurones. Similar activity in rat neocortex requires the participation of glutamatergic synaptic transmission (Garaschuk et al. 2000), and our results suggest that gap junctions may also be involved. Gap junctions are prominent among dividing neuronal precursors (Lo Turco & Kriegstein, 1991) and although dye coupling declines as neurones migrate into the cortical plate, it is still significant in the neonatal cortex (Connors et al. 1983; Peinado et al. 1993). Our data therefore suggest that the sodium current present at E16 is sufficient to drive activity among neurones, but not to synchronize it. The increase in INa density that occurs just before P0 may increase transmitter release by increasing action potential amplitude or by allowing more efficient propagation of somatal action potentials to presynaptic terminals. It may also increase electrical communication among neurones by increasing current flow across gap junctions or propagation of the action potential to the sites of gap junctional communication. Either or both effects would tend to augment synchronization of activity.

Synchronous activity disappears shortly after P0 even though INa density does not decrease. The large decrease in input resistance that occurs at this time, however, may reverse the effects of large Na+ currents on synchronization. Indeed, the developmental decrease in input resistance in cortical neurones has been shown to shorten the length constant and thus isolate the soma from other cellular compartments, where gap junctional communication may be located (Peinado et al. 1993; Zhu, 2000). In addition, decreased membrane resistance would reduce the neurone's response both to incoming synaptic activity and to its own Na+ current.

Whatever the mechanism responsible for the development of synchronicity, it most likely involves differential communication between localized clusters of neurones and the general population. Our data indicate that in many P0 slices in which global synchronous activity does not occur (possibly due to damage induced by tissue slicing or other experimental procedures), there is often still synchronous activity among localized clusters of neurones. This suggests that local communication is more robust than longer-range communication, and that the development of global synchronicity involves bringing multiple local synchronized clusters into synchrony.

Relationship to our previous work on the development of excitability in cortical neurones

Our previous data (Picken-Bahrey & Moody, 2003b) showed that many cortical neurones do not show substantial repetitive firing ability until P0–P2, and have very limited action potential generation at E16. There are several possible ways to reconcile these findings with our present results. It is possible that the spontaneous [Ca2+]i transients we record do not result at all from bursts of action potentials, or even from full-size action potentials. It is also possible that synchronous activity is driven by a subset of pacemaker neurones (see, e.g. Voigt et al. 2001) different from the deep-layer pyramidal neurones that we concentrated on in our electrophysiological experiments. Indeed, some neurones, such as subplate neurones and Cajal-Retzius cells, do appear to develop high density Na+ currents and repetitive firing ability earlier than other cortical neurones (Luhmann et al. 2000). Another likely possibility is that the current clamp results we obtained from neurones at room temperature in our previous experiments do not reflect their firing abilities at physiological temperatures. We have found that the spontaneous [Ca2+]i transients described here do not occur at temperatures below 26–30°C, and thus the ability to generate repetitive action potentials may be degraded at low temperatures. Finally, the firing ability of neurones in response to square current pulses may not give an accurate indication of their activity in response to more complex synaptic inputs (see Reinker et al. 2004; Schreiber et al. 2004).

At all the stages we examined, cortical neurones also generated large, TTX-resistant [Ca2+]i transients that were not synchronous across large populations (Fig. 8). These large transients occurred much less frequently than the small synchronous ones and unlike the small transients, increased in frequency after P0 and were not synchronous in large populations of cells. We do not know whether the TTX resistance of these large transients reflects their origin in the spontaneous release of intracellular Ca2+, or the spontaneous occurrence of purely Ca2+-dependent electrical activity, or both.

Possible factors affecting the occurrence of spontaneous activity in neonatal brain slices

In our experiments, 75% of slices showed spontaneous synchronous activity and within that sample, the number of imaged cells in a given field of view participating in the activity ranged from 20% to 100%. Some of this variability is likely to reflect injury to the tissue during preparation of the slices. Indeed, it is our experience that activity is quite sensitive to details of slice preparation, handling, and to the quality of the recording conditions. The fact that application of TEA induces highly synchronous activity in 100% of slices, even those that show no activity at all under control conditions, implies that the physical synaptic circuitry supporting synchronization is present in all slices at P0. Slice preparation may, however, depress either the strength of synaptic interactions or the intrinsic excitability of the neurones, both factors that would be overcome by the effects of TEA. Although we have no evidence for a pacemaker region for the activity, it remains possible that connections between such a pacemaker and the rest of the cortex are particularly sensitive to experimental conditions and slice preparation. A final possible source of variability is in the developmental timing of the activity. Differences in the developmental stage of animals at the time of birth or differences between animals in the stages at which activity occurs in vivo may also explain the variability in our results.

Comparison with spontaneous, synchronized activity in other developing CNS structures

The activity we have reported here in mouse cortex is similar in amplitude, frequency, and overall synchronicity to that reported by Garaschuk et al. (2000) in rat neocortex. In rat, the frequency of spontaneous transients decreases over the first postnatal week in some cortical regions, but increases in others, whereas in mouse we see a general decline in occurrence of activity in all regions postnatally. We have not examined mouse cortical slices for directional propagation of spontaneous [Ca2+]i transients, as was reported in rat. The activity in both rat and mouse cortex at these stages is lower in frequency than similar spontaneous activity in hippocampus (Strata et al. 1995; Garaschuk et al. 1998, 2000; Ben-Ari, 2001).

One of the few other structures in which the developmental emergence of spontaneous Ca2+ transients has been analysed is motor neurones of mouse brainstem cranial nerve nuclei (Gust et al. 2003). In these cells, spontaneous [Ca2+]i transients are asynchronous when they first emerge as they are in our experiments. In brainstem, however, the transients are not TTX sensitive at these early stages, and develop TTX sensitivity in concert with synchronicity over the subsequent 2 days. In cortex, the activity that occurs at E16, although generally asynchronous, is TTX sensitive. This implies that in brainstem, the ionic basis of spontaneous activity changes rapidly during early development as it become synchronous, whereas in cortex it does not.

Possible developmental significance of spontaneous, synchronous activity

The developmental function of widespread, synchronous [Ca2+]i transients must be consistent with both their occurrence in a narrow window of developmental time near birth and the very broad regions over which they occur. Several processes occurring at these stages are likely candidates for regulation by this kind of activity. Imaging of living neurones in mouse neocortex indicates very active axonal growth cone filopodia at P0, followed a few days later by similar filopodial activity in dendrites (Portera-Cailliau et al. 2003). Since axonal and dendritic outgrowth are activity-dependent processes (Catalano & Shatz, 1998; Ming et al. 2001; Wong & Ghosh, 2002), global spontaneous activity in cortical neurones at these stages could regulate the projection of cortical axons to cortical and subcortical targets as well as the arborization of cortical dendritic trees. If widespread cortical activity occurs in direction waves, as was reported by Garaschuk et al. (2000), it could also serve to identify the location of origin of cortical efferent axons to their target cells, much as retinal waves of activity do for geniculate target neurones (Wong, 1999).

The highly synchronous nature of the activity could serve to activate coincidence-detection mechanisms in emerging synaptic circuitry, and thus in a process similar to long-term potentiation, might strengthen weak or silent synaptic connections among cortical neurones (see Hanse et al. 1997; Liao et al. 2001; Kasyanov et al. 2004).

A final possible important role of widespread activity might be to regulate the maturation of intrinsic ion channel properties in cortical neurones. There are many reported cases of activity-dependent ion channel development (Linsdell & Moody, 1994, 1995; Liu & Kaczmarek, 1998; Golowasch et al. 1999; Grosse et al. 2000), and in some cases the transition out of embryonic patterns of ion channel expression that mediate spontaneous activity to the mature expression patterns is regulated by the spontaneous activity itself (Desarmenien & Spitzer, 1991; Dallman et al. 1998). The mature firing properties of cortical neurones begin to emerge over the first several postnatal days (see Massengill et al. 1997), and it is likely that spontaneous activity such as we have reported here may play a role in that process as well as others.

Acknowledgments

This work was supported by grants from the NIH and from the University of Washington Royalty Research Fund to W.J.M., and by an NSF grant to M.M.B. We acknowledge the participation of Juliane Gust in the extracellular recording experiments.

References

  1. Ben-Ari Y. Developing networks play a similar melody. Trends Neurosci. 2001;24:353–360. doi: 10.1016/s0166-2236(00)01813-0. [DOI] [PubMed] [Google Scholar]
  2. Blochl A, Thoenen H. Characterization of nerve growth factor (NGF) release from hippocampal neurons: evidence for a constitutive and an unconventional sodium-dependent regulated pathway. Eur J Neurosci. 1995;7:1220–1228. doi: 10.1111/j.1460-9568.1995.tb01112.x. [DOI] [PubMed] [Google Scholar]
  3. Catalano SM, Shatz CJ. Activity-dependent cortical target selection by thalamic axons. Science. 1998;281:559–562. doi: 10.1126/science.281.5376.559. [DOI] [PubMed] [Google Scholar]
  4. Connors BW, Bernardo LS, Prince DA. Couping between neurons of the developing rat neocortex. J Neurosci. 1983;3:773–782. doi: 10.1523/JNEUROSCI.03-04-00773.1983. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Dallman JE, Davis AK, Moody WJ. Spontaneous activity regulates Ca-activated K current expression in developing ascidian muscle. J Physiol. 1998;511:683–693. doi: 10.1111/j.1469-7793.1998.683bg.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Dallman JE, Dorman J, Moody WJ. Action potential waveform voltage clamp shows significance of different Ca2+ channel types in developing ascidian muscle. J Physiol. 2000;524:375–386. doi: 10.1111/j.1469-7793.2000.t01-1-00375.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Desarmenien MG, Spitzer NC. Role of calcium and protein kinase C in development of the delayed rectifier potassium current in Xenopus spinal neurons. Neuron. 1991;7:797–805. doi: 10.1016/0896-6273(91)90282-5. [DOI] [PubMed] [Google Scholar]
  8. Garaschuk O, Hanse E, Konnerth A. Developmental profile and synaptic origin of early network oscillations in the CA1 region of rat neonatal hippocampus. J Physiol. 1998;507:219–236. doi: 10.1111/j.1469-7793.1998.219bu.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Garaschuk O, Linn J, Eilers J, Konnerth A. Large-scale oscillatory calcium waves in the immature cortex. Nat Neurosci. 2000;3:452–459. doi: 10.1038/74823. [DOI] [PubMed] [Google Scholar]
  10. Golowasch J, Abbott LF, Marder E. Activity-dependent regulation of potassium currents in an identified neuron of the stomatogastric ganglion of the crab Cancer borealis. J Neurosci. 1999;19:RC33. doi: 10.1523/JNEUROSCI.19-20-j0004.1999. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Greaves AA, Davis AK, Dallman JE, Moody WJ. Coordinated modulation of Ca and K currents during ascidian muscle development. J Physiol. 1996;497:39–52. doi: 10.1113/jphysiol.1996.sp021748. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Grosse G, Draguhn A, Hohne L, Tapp R, Veh RW, Ahnert-Hilger G. Expression of Kv1 potassium channels in mouse hippocampal primary cultures: development and activity-dependent regulation. J Neurosci. 2000;20:1869–1882. doi: 10.1523/JNEUROSCI.20-05-01869.2000. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Gu X, Olson EC, Spitzer NC. Spontaneous neuronal calcium spikes and waves during early differentiation. J Neurosci. 1994;14:6325–6335. doi: 10.1523/JNEUROSCI.14-11-06325.1994. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Gust J, Wright JJ, Pratt EB, Bosma MM. Development of synchronized activity of cranial motor neurons in the segmented embryonic mouse hindbrain. J Physiol. 2003;550:123–133. doi: 10.1113/jphysiol.2002.038737. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Hanse E, Durand GM, Garaschuk O, Konnerth A. Activity-dependent wiring of the developing hippocampal neuronal circuit. Semin Cell Dev Biol. 1997;8:35–42. doi: 10.1006/scdb.1996.0119. [DOI] [PubMed] [Google Scholar]
  16. Kasyanov AM, Safiulina VF, Voronin L, Cherubini E. GABA-mediated giant deplolarizing potentials as coincidence detectors for enhancing synaptic efficacy in the developing hippocampus. Proc Natl Acad Sci U S A. 2004;101:3967–3972. doi: 10.1073/pnas.0305974101. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Komuro H, Rakic P. Selective role of N-type calcium channels in neuronal migration. Science. 1992;257:806. doi: 10.1126/science.1323145. [DOI] [PubMed] [Google Scholar]
  18. Komuro H, Rakic P. Orchestration of neuronal migration by activity of ion channels, neurotransmitter receptors, and intracellular calcium fluctuations. J Neurobiol. 1998;37:110–130. [PubMed] [Google Scholar]
  19. Liao D, Scannevin RH, Huganir R. Activation of silent synapses by rapid activity-dependent synaptic recruitment of AMPA receptors. J Neurosci. 2001;21:6008–6017. doi: 10.1523/JNEUROSCI.21-16-06008.2001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Linsdell P, Moody WJ. Na channel misexpression accelerates K channel development in embryonic Xenopus laevis skeletal muscle. J Physiol. 1994;480:405–410. doi: 10.1113/jphysiol.1994.sp020370. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Linsdell P, Moody WJ. Electrical activity and calcium influx regulate ion channel development in embryonic Xenopus skeletal muscle. J Neurosci. 1995;15:4507–4514. doi: 10.1523/JNEUROSCI.15-06-04507.1995. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Liu SJ, Kaczmarek LK. Depolarization selectively increases the expression of the Kv3.1 potassium channel in developing inferior colliculus neurons. J Neurosci. 1998;18:8758–8769. doi: 10.1523/JNEUROSCI.18-21-08758.1998. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Lo Turco JJ, Kriegstein AR. Clusters of coupled neuroblasts in embryonic neocortex. Science. 1991;252:563–566. doi: 10.1126/science.1850552. [DOI] [PubMed] [Google Scholar]
  24. Luhmann HJ, Reiprich RA, Hanganu I, Kilb W. Cellular physiology of the neonatal rat cerebral cortex: intrinsic membrane properties, sodium and calcium currents. J Neurosci Res. 2000;62:574–584. doi: 10.1002/1097-4547(20001115)62:4<574::AID-JNR12>3.0.CO;2-0. [DOI] [PubMed] [Google Scholar]
  25. Massengill JL, Smith MA, Son DI, O'Dowd DK. Differential expression of K4-AP currents and Kv3.1 potassium channel transcripts in cortical neurons that develop distinct firing phenotypes. J Neurosci. 1997;17:3136–3147. doi: 10.1523/JNEUROSCI.17-09-03136.1997. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Meister M, Wong RO, Baylor DA, Shatz CJ. Synchronous bursts of action potentials in ganglion cells of the developing mammalian retina. Science. 1991;252:939–943. doi: 10.1126/science.2035024. [DOI] [PubMed] [Google Scholar]
  27. Meyer-Franke A, Wilkinson GA, Kruttgen A, Hu M, Munro E, Hanson MG, Jr, Reichardt LF, Barres BA. Depolarization and cAMP elevation rapidly recruit TrkB to the plasma membrane of CNS neurons. Neuron. 1998;21:681–693. doi: 10.1016/s0896-6273(00)80586-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Ming G, Henley J, Tessier-Lavigne M, Song H, Poo M. Electrical activity modulates growth cone guidance by diffusible factors. Neuron. 2001;29:441–452. doi: 10.1016/s0896-6273(01)00217-3. [DOI] [PubMed] [Google Scholar]
  29. O'Donovan MJ. The origin of spontaneous activity in developing networks of the vertebrate nervous system. Curr Opin Neurobiol. 1999;9:94–104. doi: 10.1016/s0959-4388(99)80012-9. [DOI] [PubMed] [Google Scholar]
  30. O'Leary DDM, Ruff NL, Dyck RH. Development, critical period plasticity, and adult reorganizations of mammalian somatosensory systems. Curr Opin Neurobiol. 1994;4:535–544. doi: 10.1016/0959-4388(94)90054-x. [DOI] [PubMed] [Google Scholar]
  31. Peinado A, Yuste R, Katz LC. Extensive dye coupling between rat neocortical neurons during the period of circuit formation. Neuron. 1993;10:103–114. doi: 10.1016/0896-6273(93)90246-n. [DOI] [PubMed] [Google Scholar]
  32. Picken-Bahrey HL, Moody WJ. Voltage-gated currents, dye and electrical coupling in the embryonic mouse neocortex. Cereb Cortex. 2003a;13:239–251. doi: 10.1093/cercor/13.3.239. [DOI] [PubMed] [Google Scholar]
  33. Picken-Bahrey HL, Moody WJ. Early development of voltage-gated ion currents and firing properties in neurons of the mouse cerebral cortex. J Neurophysiol. 2003b;89:1761–1773. doi: 10.1152/jn.00972.2002. [DOI] [PubMed] [Google Scholar]
  34. Portera-Cailliau C, Pan DT, Yuste R. Activity-regulated dynamic behavior of early dendritic protrusions: evidence for different types of dendritic filopodia. J Neurosci. 2003;23:7129–7142. doi: 10.1523/JNEUROSCI.23-18-07129.2003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Reinker S, Puil E, Miura RM. Membrane resonance and stochastic resonance modulate firing patterns of thalamocortical neurons. J Comput Neurosci. 2004;16:15–25. doi: 10.1023/b:jcns.0000004838.67584.77. [DOI] [PubMed] [Google Scholar]
  36. Schreiber S, Fellous JM, Tiesinga P, Sejnowski TJ. Influence of ionic conductances on spike timing reliability of cortical neurons for suprathreshold rhythmic inputs. J Neurophysiol. 2004;91:194–205. doi: 10.1152/jn.00556.2003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Shatz CJ, Stryker MP. Prenatal TTX infusion blocks segregation of retinogeniculate afferents. Science. 1988;242:87–89. doi: 10.1126/science.3175636. [DOI] [PubMed] [Google Scholar]
  38. Shieh PB, Ghosh A. Molecular mechanisms underlying activity-dependent regulation of BDNF expression. J Neurobiol. 1999;41:127–134. [PubMed] [Google Scholar]
  39. Sisk R, Moody WJ. Spontaneous, synchronized electrical activity in neonatal mouse cortical neurons. Abstr Soc Neurosci. 2003 676.10, 2003 Abstract viewer/itinerary Planner, Washington, DC. [Google Scholar]
  40. Spitzer NC. Activity-dependent neuronal differentiation prior to synapse formation: the functions of calcium transients. J Physiol (Paris) 2002;96:73–80. doi: 10.1016/s0928-4257(01)00082-1. [DOI] [PubMed] [Google Scholar]
  41. Strata F, Sciancalepore M, Cherubini E. Cyclic AMP-dependent modulation of giant depolarizing potentials by metabotropic glutamate receptors in the rat hippocampus. J Physiol. 1995;489:115–125. doi: 10.1113/jphysiol.1995.sp021035. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Svoboda R, Linares AE, Ribera AB. Activity regulates programmed cell death of zebrafish Rohon-Beard neurons. Development. 2001;128:3511–3520. doi: 10.1242/dev.128.18.3511. [DOI] [PubMed] [Google Scholar]
  43. Voigt T, Opitz T, de Lima AD. Synchronous oscillatory activity in immature cortical network is driven by GABAergic preplate neurons. J Neurosci. 2001;21:8895–8905. doi: 10.1523/JNEUROSCI.21-22-08895.2001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Wenner P, O'Donovan MJ. Mechanisms that initiate spontaneous network activity in the developing chick spinal cord. J Neurophysiol. 2001;86:1481–1498. doi: 10.1152/jn.2001.86.3.1481. [DOI] [PubMed] [Google Scholar]
  45. Wong ROL. Retinal waves and visual system development. Ann Rev Neurosci. 1999;22:29–47. doi: 10.1146/annurev.neuro.22.1.29. [DOI] [PubMed] [Google Scholar]
  46. Wong ROL, Ghosh A. Activity-dependent regulation of dendritic growth and patterning. Nat Rev Neurosci. 2002;3:803–812. doi: 10.1038/nrn941. [DOI] [PubMed] [Google Scholar]
  47. Zhu JJ. Maturation of layer 5 neocortical pyramidal neurons: amplifying salient layer I and layer 4 inputs by Ca2+ action potentials in adult rat tuft dendrites. J Physiol. 2000;526:571–587. doi: 10.1111/j.1469-7793.2000.00571.x. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from The Journal of Physiology are provided here courtesy of The Physiological Society

RESOURCES