| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
REPORT
1Department of Physiology and Biophysics, Georgetown University, Washington, DC; 2Institute of Neurosciences, University Miguel Hernandez, Alicante, Spain; and 3Department of Fisiologia, University of La Laguna, Tenerife, Spain
Submitted 28 March 2005; accepted in final form 16 September 2005
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
; Katz and Shatz 1996). Such activity has been found in many parts of the CNS, including retina (Maffei and Galli-Resta 1990; Meister et al. 1991; Wong 1999), hippocampus (Ben-Ari et al. 1989; Menendez de la Prida et al. 1998), neocortex (Garaschuk et al. 2000; Lippe 1994; McCormick 2001; Yuste et al. 1992), thalamus (Andersen et al. 1996), spinal cord (O'Donovan et al. 1994; Spitzer and Gu 1997), and embryonic CNS (Momose-Sato et al. 2003).
In rodent hippocampus, synchronous neuronal discharges known as giant depolarizing potentials (GDPs) appear during the first postnatal week (Ben-Ari et al. 1989; Khalilov et al. 1997; Leinekugel et al. 1998; Menendez de la Prida et al. 1996). GDPs have been well characterized by intracellular recordings, as depolarizations of 20–50 mV for 300–500 ms, with superimposed action potentials and an occurrence rate of 3–12 GDPs/min. The depolarization of the neurons is thought to be generated by the synchronized activation of interneurons and pyramidal cells through GABAA, AMPA, and N-methyl-D-aspartate (NMDA) receptors (Ben-Ari et al. 1989; Bolea et al. 1999; Gaiarsa et al. 1991; Khazipov et al. 1997). GABA mediate circuits play an important role in GDPs. Both spontaneous and evoked GDPs disappear when GABAA receptors are blocked (Ben-Ari et al. 1989; Khalilov et al. 1999; Menendez de la Prida et al. 1996). Experimental evidence showed that, during early postnatal development, GABA responses can be depolarizing and excitatory in some systems because of a high intracellular [Cl–] (Chen et al. 1996; Cherubini et al. 1991; Owens et al. 1996; Serafini et al. 1995).
In addition to GABAA receptors, glutamate receptors also contribute to the initiation and sustaining of GDP events (Ben-Ari et al. 1989; Bolea et al. 1999; Gaiarsa et al. 1991; Khazipov et al. 1997). GDPs can be blocked by 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX) and 1-(4-aminophenyl)-4-methyl-7,8-methylenedioxy-5H-2,3-benzodiazepine (GYKI 53655) (Bolea et al. 1999; Gaiarsa et al. 1991), suggesting that AMPA-mediated depolarization is essential to GDP events. NMDA receptors may also contribute to the GDP as observed calcium influx through NMDA- and voltage-dependent calcium channels (Garaschuk et al. 1998; Leinekugel et al. 1995, 1997). However, D-2-amino-5-phosphonopentanoic acid (APV) only partially blocks the occurrence of GDPs (Bolea et al. 1999), suggesting that NMDA receptors and subsequent calcium influx are not essential.
Spontaneous GDPs can occur in isolated CA3, CA1, or the hilus (Menendez de la Prida et al. 1998), suggesting that they emerge as a property of local circuits in all hippocampal structures. Recently, GDPs have also been recorded in vivo in rat pups, where they occur during immobility periods, sleep, and feeding (Leinekugel et al. 2002) at similar frequency as in acute slices (6–20 GDPs/min).
GDPs may play a role in the control of functional synapse number (Lauri et al. 2003), as well as "unsilencing" silent connections and strengthening those with low probability of transmitter release (Gasparini et al. 2000). Increased GABAA receptor activation seems to accelerate the switch of GABAergic responses from excitatory to inhibitory (Ganguly et al. 2001). Neurons rarely fire between GDP bursts at this age, leaving GDPs as the main source of correlated neuronal activity. Understanding the spatiotemporal pattern of GDPs is crucial to understanding when and where the neurons will fire together and how such synchronized population activation refine the neuronal connectivity in this developing stage.
The spatiotemporal pattern of the GDP has not been examined by imaging methods. There may be two possible spatiotemporal regimens. If the GDP initiates from a point pacemaker and propagates through local connections, the activity would progressively invade the tissue and appear as a propagating wave. If the propagation follows the trisynaptic circuit between dentate gyrus (DG), CA1, and CA3, the excitation would appear almost simultaneously (with about 1 synaptic delay) in the CA3 and CA1. Obviously the two possible regimens would provide different spatial synchrony in the developing hippocampus. In this report, we use voltage-sensitive dye (VSD) imaging to visualize the initiation and propagation of GDPs in hippocampal slices from neonatal rats. Our data show that the propagation of GDPs follows the first regimen: GDP initiated from a small focus and progressively propagated as a wave front. Such spatiotemporal pattern may contribute to refining of hippocampal circuits by providing a localized spatial synchrony.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Hippocampal slices were obtained from P2–P6 Sprague-Dawley rats of both sexes. The animals were killed following National Institutes of Health guidelines, deeply anesthetized with halothane, and quickly decapitated. The whole brain was immersed in cold (0–4°C) artificial cerebrospinal fluid (ACSF) containing (in mM) 126 NaCl, 3.5 KCl, 2 CaCl2, 1.3 MgSO4, 1.25 NaH2PO4, 26 NaHCO3, and 10 dextrose and saturated with 95% O2-5% CO2 (pH = 7.4). The brain was chilled for 90 s, and the hippocampus was isolated for slicing.
Transverse hippocampal slices (400 µm thick) were cut with a brain tissue slicer (Stoelting) and allowed to recover at room temperature for 
1 h before staining. Slices were perfused with ACSF containing 0.02 mg/ml of the oxonol VSD RH 479 (1st synthesized by R. Hildesheim and A. Grinvald and kindly provided by Dr. L. Loew, University of Connecticut, Farmington, CT, as JPW 1131) for 30 min. The stained preparation was perfused with dye-free ACSF at room temperature for 1 h before optical imaging. During the experiment, the slice was continuously perfused with ACSF in a submerged chamber on the microscope stage.
VSD imaging
Optical imaging was performed with a 124-element photodiode array (Centronics, Newbury Park, CA) at a frame rate of 1,000 frames/s. The image of the preparation was projected on the array, such that each detector received light from a 42 x 42-, 84 x 84-, or 168 x 168-µm2 area of the slice, depending on the objective used (x40, x20, or x10, respectively). The photocurrent of dye-related absorption signals (705 nm) from each photodetector was individually amplified through a two-stage amplifier. The first stage performed a current to voltage conversion with a feedback resistor of 50 M
. The signals were amplified and filtered by a second stage amplifier before digitizing. This second stage provided a voltage gain of 200, high-pass filtering with a 100-ms time constant, and a four-pole Bessel analog low-pass filter with a 300-Hz corner frequency. [The commercial version of the recording apparatus (with 464 detectors) is available as NeuroPDA from RedshirtImaging; www.redshirtimaging.com]. The recording apparatus allows low dark noise (10–6 of the illumination intensity), large effective dynamic range (>19 bits), and fast sampling rate (1,000 frames/s). With RH479 and absorption measurement, the optical signal size (dI/I) for the GDP was about 5 x 10–4–2 x 10–3. This value was calculated by dividing the measured signal by the resting light intensity. In trace displays (Figs. 1 and 2, B and C), dI/I was used to show the amplitude of the optical signal. In pseudocolor displays (Fig. 2D and all contour displays), we did not calculate dI/I; instead, the measured signal on each detector was normalized to its own maximum amplitude. The resting photocurrent on the diode was 
100 nA. At this illumination intensity the signal-to-noise ratio was 10, so signal averaging was not necessary. Optical recording trials were 2 s long for the evoked GDPs and 8–40 s for spontaneous GDPs. On each slice, we recorded 4–20 trials (with a total exposure of <160 s). Phototoxicity was not a concern with this level of exposure time and intensity (Jin et al. 2002). For additional details about the apparatus and the recording methods, see Jin et al. 2002; Kosmidis et al. 2004; Wu and Cohen 1993.
|
|
Electrical recording
Tungsten (epoxy-coated) microelectrodes with tip resistance of 75 k (FHC, Bowdoinham, ME) were used for recording local field potentials simultaneously with the optical recordings. The electrodes were carefully inserted into cortical tissue so that the tissue surrounding it produced normal VSD signals. The field potential signals were amplified 1,000 times and band-pass filtered between 0.1 and 400 Hz by a Brownlee Precision 440 amplifier. During imaging experiments, the field potential was digitized simultaneously with VSD signals. The same tungsten microelectrode was also used for stimulation to trigger GDP events. Stimuli consisted of rectangular pulses of 10–30 V and 100 µs in duration.
Data analysis
The optical data were analyzed using the program NeuroPlex (Redshirtimaging, Fairfield, CT). NeuroPlex displays the data in the form of traces for numerical analysis (Fig. 2B) or pseudocolor images for visualizing the spatiotemporal pattern (Fig. 2D). The digitized data were further filtered digitally by NeuroPlex [low-pass, 300 Hz (Butterworth) and high-pass, 1.2 Hz (RC filter)] before displaying. The pseudocolor display was made using variable scaling, where the data from each diode are first normalized to their own maximum and then each normalized data point is assigned a color according to a linear color scale (0 = deep blue and 1 = red). This kind of scaling best preserves the timing information of the signals. The display was either in pixel (Fig. 2D, top row) or contour mode (Fig. 2D, bottom row). For analysis of the propagation activity, contour mode was chosen. Contour mode low-pass filters the raw data spatially, so that areas of equal signal amplitude are assigned the same color, and the amplitude difference between neighboring diodes is interpolated. To determine the initiation point of the discharges, pixel mode can better show which detectors depolarize first.
To calculate the propagation velocity between two locations, we first measured the time for the GDP to reach 50% of the peak amplitude at the two locations (Fig. 3C); the propagation velocity is calculated as the distance between the two locations divided by the difference of the two times.
|
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
In our experimental arrangement, the VSD signal is wavelength-dependent light absorption. The light absorption increased at 705 nm and decreased at 650 nm in association with the membrane depolarization of the cortical neuronal population. This wavelength-dependent signal is referred to as optical signal or VSD signal (Cohen et al. 1974). This VSD signal occurred simultaneously with the local field potential signals of GDPs (Fig. 1), and could be blocked by GABAA and glutamate antagonists (data not shown). Therefore we presume that the VSD signals reflect the population neuronal depolarization of the GDPs. Local field potential recordings from nonstained and nonimaged slices showed similar amplitude, duration, and initiation frequency, indicating that the dye staining and the light exposure during imaging did not have noticeable effects. The optical signal has a smooth waveform, whereas local field potential signal has more high-frequency components (Fig. 1). This might be because the optical signal is the sum of the membrane depolarization of a population (in an area of 42 x 42 µm2), whereas the local field potential electrode collects both spikes and local current because of depolarization and repolarization.
Propagation of GDP waves
The propagation of the optical signals during an evoked GDP in the CA3 area is shown in Fig. 2. The GDP was evoked by an electrical stimulation in the hilus (outside the field of view; Fig. 2A, stim). All of the optical detectors recorded optical signals of the evoked GDP, except those measuring light from an area at the border or outside the slice (Fig. 2B, dashed line), which were excluded from analysis. Signal-to-noise ratio of the optical signals is large enough for making maps of propagation pattern and velocity without averaging (Figs. 1 and 2B).
Pseudocolor images derived from the amplitude map of the optical signals (Fig. 2D, top row) showed a wave-like pattern of propagation progressively invading the previously resting areas. In the example shown here, the wave of the activity came from the top of the recording area (closer to the stimulation point) and propagated to the bottom. Although images were acquired every 1 ms, for the sake of simplicity, we only show an image every 5 ms. Contour mode of the pseudocolor map allows more detailed analysis of the propagation pattern (Fig. 2D, bottom row). Isoamplitude lines, or the borderline between two given colors, are used to label the propagation front of the waves (Fig. 2D, bottom row, black lines). By drawing the isoamplitude lines at every millisecond on the map, the propagation course of the GDP wave can be obtained (Fig. 2E). The isoamplitude lines show regions having the same level of excitation at the time, i.e., that are activated at the same time. Conversely, regions located on different isoamplitude lines have lower degrees of synchrony. The map indicates a striking characteristic of the propagation pattern—directional selectivity that is correlated to the CA3 anatomic structure (Fig. 2E). The isoamplitude lines are perpendicular to the major axis of the hippocampus, the pyramidal layer (Fig. 2E, broken lines). This strongly suggests a higher degree of synchrony along the axis of imaginary column of pyramidal neurons, which includes the soma and dendritic tree of pyramidal cells. The gaps between adjacent contour lines are proportional to the propagation velocity; the wider the gap, the faster the wave travels at this location. Contour lines in Fig. 2E show that the wave gradually accelerated when propagating along the major axis in CA3.
Different propagation patterns in hilus and CA3
To further analyze the relationship between anatomic structure and GDP propagation, we examined the propagation patterns in different regions of the developing hippocampus. In Fig. 3, we show an example of the propagation patterns of GDPs evoked in the hilus. The activity propagated from the left, the initiation point in the hilus, to the right, toward CA3 (see location of stimulation electrode in Fig. 3A). The isoamplitude contour lines show a different pattern in the hilus compared with that in CA3 (Fig. 3B). In the hilus, the propagation was slower (lines are closer) and more convoluted. The propagation velocity became higher in CA3, and the wave front became less convoluted, running perpendicular to the pyramidal layer. These results suggest that the propagation pattern of GDPs is influenced by cellular architecture of the hippocampus. In Fig. 3C, optical signals from three groups of optical detectors (blue, red, and green squares in Fig. 3, A and B) were normalized to their maximum and plotted. The green squares were located along the axis of an imaginary pyramidal neuron in CA3 and had the best level of synchronization; the activation at these locations reached 50% of the maximum almost simultaneously (Fig. 3C). The blue squares were located in the hilus area and showed the least synchrony; the time to reach 50% of activation differed by 12 ms. The red squares were at the hilus–CA3 border and had an intermediate level of synchrony.
These results suggest that the shape of the wave front is highly correlated to the cellular organization in hilus and CA3. The propagating pattern in CA1 is similar to that of CA3 but with different velocity.
Propagation velocity
Observations in Figs. 2E and 3B suggest that the propagation of GDPs becomes faster along the path in hilus and CA3. To test if this is true across the whole propagation path, we used a low power objective (x10) to image hilus, CA3, and CA1 simultaneously (Fig. 4A). In 28 events from two different slices, the GDPs propagated across CA1 in the same pattern as that in CA3, with isoamplitude lines perpendicular to the pyramidal layer (Fig. 4A, contour lines). However, as a general tendency, the velocity in CA1 was higher than that in CA3. In the front edge contour map, it is clear that the distance between isoamplitude lines increases in CA1, suggesting an increase in the propagation velocity. A row of detectors along the pyramidal cell layer in CA3 and CA1 was selected for calculating the propagation velocity in the example presented. In Fig. 4, eight detectors were selected. The optical signals were normalized to the maximal amplitude on each detector, and the time for the signal to reach its 50% of the maximal amplitude was marked as the GDP arrival time at that location (Fig. 4B). The propagation velocity was defined as the distance between the two detectors divided by the difference of the two arrival times. In the GDP event shown in Fig. 4B, detectors 5 and 6 had similar arrival times, indicating a high velocity. The propagation velocity is significantly different in three hippocampal subfields: it was slowest in hilus, 1–5 mm/s; faster in CA3, 2.8–60 mm/s (with large local variations); fastest in CA1, 98–167 mm/s. GDPs evoked in CA1 by microshocks propagated backward to the CA3 and then to hilus. The velocity of back propagation was similar to that of forward propagation: slowest in hilius, faster in CA3, and fastest in CA1.
|
In the previous sections, we have examined the variation of the GDP propagation related to anatomical organization in three hippocampal areas. Here we examine variations of the propagation velocity influenced by dynamic factors, i.e., variations of the velocity from trial to trial in a given area of the same slice.
For this purpose, we imaged 17 evoked GDPs in the same slice and examined the propagation in CA3 (Fig. 5). The GDP episodes were evoked by an extracellular electrode placed in the hilus, outside the field of view (Fig. 5A). In these GDP episodes evoked by stimuli of identical intensity and interstimulus interval, we found that some episodes were significantly faster or slower than the average. The contour maps of two example trials, Fast and Slow, are shown in Fig. 5B to demonstrate overall variability in the propagation velocity.
|
30 mm/s (e.g., trials 2 and 3), which is about twice as fast as the slow trials (trials 6, 16, and 17). The average velocity appeared to be independent of the initial velocity at the beginning of the propagation path (e.g., trial 13 has a faster velocity than trial 3 at the beginning of the path, but a slower averaged speed overall), suggesting that the variation is not caused by the initiation of GDP. Another type of trial-to-trial variation is the local velocity. Along the propagation path, waves in some regions are faster than in the other places (e.g., the red areas surrounding the broken line in trials 1–3, 5, and 15). The velocity varies largely (10–60 mm/s). The fast regions are not fast in all trials, because slower velocity is seen in trials 6, 10, 14, and 16. This strongly suggests that, in a given anatomical structure, dynamic factor(s) can significantly change the velocity.
The averaged propagation profile of all trials is shown in the bottom panel of Fig. 5C. This averaged profile shows an overall tendency of gradual increase in propagation velocity in the CA3 area, consistent with the data shown in Figs. 2E, 3C, and 4C. However, the propagation profile of individual events may differ from this general tendency. In trials 6 and 16, the velocity was faster in the first portion of CA3 and then became slower. Even the locations with a faster velocity on average (Fig. 5C, arrows) can sometimes show a much slower velocity. Overall, the large trial-to-trial variations in propagation velocity suggest that dynamic factors contribute significantly over the spreading of the GDP. Large trial-to-trial variations in propagation velocity were observed in all slices from animals of P2–P6. In each slice, we measured 10 GDP events evoked in hilus (total of 114 trials from 10 slices). The average velocity over the entire CA3 area was 21.8 ± 5.6 mm/s. However, at different local areas, the velocity varied from 2.8 to 60 mm/s, consistent with the example shown in Fig. 5.
Initiation foci of GDPs
It has been controversial whether spontaneous GDPs originate only in the hilus (Strata et al. 1997) or also can be initiated in other hippocampal areas (Bolea et al. 1999; Menendez de la Prida et al. 1998). VSD imaging is fast enough to locate the initiation foci of spontaneous and evoked GDPs. We have observed multiple initiation foci in each slice distributed in both hilus and CA3 areas (Fig. 6). Figure 6A shows the locations of the initiation foci for seven spontaneous GDPs (black boxes). Three of them are in the CA3 pyramidal layer and the other four are in the hilus. This suggests that the rhythm of GDPs may be produced by multiple pacemakers distributed both in the hilus and the CA3. We therefore set out to learn whether evoked GDPs were initiated from the same foci that generated spontaneous GDPs. We imaged five evoked GDP events in the same slice. Each initiation focus is labeled in Fig. 6A with gray boxes. Surprisingly, the evoked GDP did not start at the tip of the electrode, and identical stimuli did not evoke GDPs from the same focus (interstimulus interval > 3 min). These foci were distributed as far as 150 µm from each other, and some were far away from the electrode tip (Fig. 6A, gray boxes). In addition, the initiation foci of spontaneous GDPs did not overlap with those of evoked ones (Fig. 6A, gray and black boxes). Figure 6, B and C, shows the front edge contour maps (1-ms isoamplitude lines) of two spontaneous and three evoked GDPs from the same slice. In all cases, the GDPs started from small foci (
) and propagated toward CA3 and CA1 (right) and the hilus (left). Note that the propagation velocity became lower when the waves enter the hilus (1–5 mm/s). The isoamplitude lines on the top left of all the contour maps in Fig. 6, B and C, are more densely arranged than those in other areas, indicating slower velocity. In another slice, we imaged the initiation of 24 spontaneous GDPs with a larger field of view (x20 objective; Fig. 6D). Again the initiation foci for spontaneous GDPs were marked as black boxes and initiation foci for evoked GDPs as gray. In this slice, all foci were in the CA3 area. There were two preferred locations within CA3,: one close to the hilus (16/24 GDPs) and the other close to CA1 (8/24). The distance between the two preferred areas was 350 µm, and the initiation foci of consecutive GDPs jumped from one to the other. In this slice, we did not see spontaneous initiation foci in the hilus. However, an electrode placed in the hilus was able to reliably evoke GDPs (gray box in Fig. 6D). When the stimulation electrode was placed in CA1, microshocks were also able to reliably evoke GDPs at the tip of the electrode, suggesting that CA1 has a potential to generate GDP events from local recurrent interactions (203 trials in 12 slices).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
; Jin et al. 2002). This technique has been used to visualize epileptiform and oscillatory activities in hippocampal and neocortical slices, and propagating waves are a common spatiotemporal characteristic in these activities (Albowitz and Kuhnt 1995; Bao and Wu 2003; Demir et al. 1998, 2000; Huang et al. 2004; Miyakawa et al. 2003; Tsau et al. 1998; Wu et al. 1999b, 2001). In this study, we found that GDP activity in developing hippocampus appears as propagating waves and that the propagation patterns are strongly correlated to anatomical structure in the CA3 and CA1 areas. Large trial-to-trial variations in the propagation velocity were found, suggesting dynamic control of GDP spatiotemporal distribution. We also show that GDPs are initiated from a small focus and that, in each slice, different GDP events may be initiated from different locations. Propagation pattern of GDP waves
We found a correlation between GDP propagation pattern and the anatomical organization in different regions of hippocampus. In the hilus, where mossy cells and interneurons are not organized in layers, the propagation of GDP waves had similar velocity in all directions, and the isoamplitude lines in the contour map were arranged in a relatively concentric pattern (Figs. 3 and 4). In CA3 and CA1 areas, the distribution of pyramidal neurons was highly organized, with the somas densely grouped in the pyramidal layer and the dendritic trees aligned toward the center of the slice. When the wave reaches the CA3 area, the progression of the wave front (isoamplitude lines) became parallel to the orientation of the pyramidal cell dendritic trees (Figs. 2E, 3B, and 4A). Neurons located within the same isoamplitude line depolarized with greater synchrony. Other reports showed that, not only do pyramidal cells fire synchronous bursts, interneurons also show this property (Menendez de la Prida and Sanchez-Andres 2000), and the firing is associated with the increase in [Ca2+]i (Garaschuk et al. 1998; Leinekugel et al. 1995). Taken together, this indicates that, in the CA3 and CA1 areas, the wave of excitation implicates regions of maximal activity associated with the pyramidal neurons, both temporally and spatially. This is similar to the developing neocortex, where the activation of muscarinic acetylcholine receptors results in waves of correlated activity that propagate parallel to the pial surface and perpendicular to the cortical lamina (Peinado 2000).
Our data also show that GDP propagation is correlated to anatomical structures but is not restricted by the classical trisynaptic hippocampal circuit. The propagation of GDPs follows the major axis of the hippocampus from dentate gyrus to CA3 then to CA1, which is compatible with the classical trisynaptic circuit (Andersen et al. 1971). However, if GDPs propagated through the trisynaptic circuits from hilus to CA1, we would expect the signal arriving at CA1 cells monosynaptically through the Schaffer collaterals before running polysynaptically. We would not see progressive wave fronts running step by step along the major axis of hippocampus. We also found that GDPs evoked in CA1 or CA3 propagated "backward" to the hilus. This backward propagation cannot be explained in terms of the trisynaptic circuit. Thus the observations in this report argue against an essential contribution of the long excitatory pathways but are in agreement with the limited functionality of the glutamatergic synapses at this early stage of development (Durand et al. 1996). However, we noticed that the long fibers in the trisynaptic circuit were more easily damaged during slicing of the brain, so our results remain to be verified in vivo. Our arguments are supported by other studies. Simultaneous electrical recordings showed that GDPs could appear first in CA1 and later in CA3 pyramidal cells (Menendez de la Prida et al. 1998). Amaral and Witter (1989), and Amaral (1993) have questioned the validity of the trisynaptic circuit in the newborn. Canepari et al. (2000) used calcium imaging to examine the GDP propagation during the first postnatal week. They found that, in the presence of kynurenic acid, which blocks the glutamate component, the GDP can still propagate toward the CA1, probably through GABA-mediated local networks (Canepari et al. 2000).
Sequential activation along the axis of pyramidal laminar may contribute to the consolidation of developing circuits by providing reduced synchrony over distance. Recently, it has been reported that the pairing of GDPs with mossy fiber (MF) stimulation induced a persistent increase in synaptic strength at MF-CA3 synapses. The potentiation was maximum when pre- and postsynaptic signals were coincident and decreased when the interval between GDPs and MF stimulation increased (Kasyanov et al. 2004). In light of this observation, we would speculate that the propagation wave of GDP would contribute to the construction of local synaptic domains.
Dynamic variations in propagation velocity
In the same slice, the propagation velocity and the pattern of the contour map varied from trial to trial. These variations were large and suggest that dynamic factors contribute significantly to the control of propagation.
We observed two types of dynamic variations in the GDP propagation: one was the variation in averaged velocity over the entire CA3 (Fig. 5C, fast and slow trials). Another was the trial-to-trial variation of local velocity (Fig. 5C), which showed a large difference. Our analysis also suggests that the differences in averaged velocity may be primarily determined by the occurrence of fast/slow regions. For example, fast trials in Fig. 5C (trials 2 and 3) had more fast regions than slow trials (trials 6 and 14). Examining the variations of individual events also suggests that the general tendency of acceleration in CA3 is caused by the increasing of occurrence rate of fast regions along the propagation path from CA3 to CA1.
Initiation of GDP
We observed that GDP started from a small point focus instead of an area of coactivation (Fig. 6). Such "point initiation" was also seen in the initiation of epileptiform events observed in disinhibited neocortex (Ma et al. 2004; Tsau et al. 1998, 1999), whereas "area initiation" (or "plateau activity" Demir et al. 2000) has been observed in rat piriform cortical slices. The synaptic mechanism for GDP initiation is different from that of epileptiform activity (Khalilov et al. 1999). In disinhibited cortex, the initiation is likely to be mediated by a recurrent excitation network with glutamate synapses (McCormick and Contreras 2001). In the developing hippocampus, both glutamatergic and GABAergic networks contribute to the recurrent excitations (Bolea et al. 1999; Chen et al. 1996; Cherubini et al. 1991). Thus the initiation of GPDs is likely to be mediated by both networks.
The hippocampal region for GDP initiation has been controversial. Earlier studies pointed to an exclusive localization in the hilus (Strata et al. 1997). Later simultaneous intracellular recordings from CA3 and CA1 pyramidal cells suggested CA3 and CA1 also can initiate GDP ( Menendez de la Prida and Sanchez-Andres 2000; Menendez de la Prida et al. 1998). In addition, isolated tissue with CA3 and CA1 fields can also generate spontaneous GDPs (Bolea et al. 1999; Garaschuk et al. 1998; Menendez de la Prida et al. 1998). In this study, we confirmed that spontaneous GDPs can originate both in the hilus and in CA3 (Fig. 6). The initiation site is not fixed but varies from event to event. Even within the hilus or CA3, it changes from one burst to the next. While we did not see spontaneous GDPs originating in the CA1 area, our optical recording trials sampled a very small fraction of time, and we would definitely miss infrequent initiations in CA1. Electrical stimulation to CA1 can initiate GDP events starting in the CA1 area (data not shown), suggesting CA1 at least has the potential to initiate GDPs.
In conclusion, GDPs initiate from a small focus and propagate to other areas as waves, similar to many cortical population neuronal events. In the developing hippocampus, both GABAergic and glutamate networks are excitatory, and either of them may mediate the initiation/propagation of GDP. Questions remain following this report. Are different propagation patterns in hilus and CA3/CA1 mediated by different neuronal networks? Further experiments are needed to understand the roles of these two networks.
|
GRANTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Present address of S. Bolea: Laboratory of Sensory Processing, Brain and Mind Institute, EPFL, Lausanne, Switzerland.
|
FOOTNOTES |
|---|
Address for reprint requests and other correspondence: J.-Y. Wu, Dept. of Physiology and Biophysics, Georgetown Univ., 3970 Reservoir Rd. NW, Washington, DC 20057-1421 (E-mail: wuj{at}georgetown.edu)
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Amaral DG. Emerging principles of intrinsic hippocampal organization. Curr Opin Neurobiol 3: 225–229, 1993.[CrossRef][Medline]
Amaral DG and Witter MP. The three-dimensional organization of the hippocampal formation: a review of anatomical data. Neuroscience 31: 571–591, 1989.[CrossRef][Web of Science][Medline]
Andersen P, Andersson SA, and Lomo T. Patterns of spontaneous rhythmic activity within various thalamic nuclei. Nature 211: 888–889, 1996.
Andersen P, Bliss TVP, and Skrede K. Lamellar organization of hippocampal excitatory pathways. Exp Brain Res 13: 222–238, 1971.[Web of Science][Medline]
Bao W and Wu JY. Propagating waves and irregular dynamics: spatiotemporal patterns of cholinergic theta oscillations in neocortex, in vitro. J Neurophysiol 90: 333–341, 2003.
Ben-Ari Y, Cherubini E, Corradetti R, and Gaiarsa JL. Giant synaptic potentials in inmature rat CA3 hippocampal neurones. J Physiol 416: 303–325, 1989.
Bolea S, Avignone E, Berretta N, Sanchez-Andres JV, and Cherubini E. Glutamate controls the induction of GABA-mediated giant depolarizing potentials through AMPA receptors in neonatal rat hippocampal slices. J Neurophysiol 81: 2095–2102, 1999.
Canepari M, Mammano F, Kachalsky SG, Rahamimoff R, and Cherubini E. GABA- and glutamate-mediated network activity in the hippocampus of neonatal and juvenile rats revealed by fast calcium imaging. Cell Calcium 27: 25–33, 2000.[CrossRef][Web of Science][Medline]
Chen G, Trombley P, and Van Den Pol AN. Excitatory actions of GABA in developing rat hypothalamic neurones. J Physiol 494: 451–464, 1996.
Cherubini E, Gaiarsa JL, and Ben-Ari Y. GABA: an excitatory transmitter in early postnatal life. Trends Neurosci 14: 515–519, 1991.[CrossRef][Web of Science][Medline]
Cohen LB, Salzberg BM, Davila HV, Ross WN, Landowne D, Waggoner AS, and Wang CH. Changes in axon fluorescence during activity: molecular probes of membrane potential. J Membr Biol 19: 1–36, 1974.[CrossRef][Web of Science][Medline]
Colom LV and Saggau P. Spontaneous interictal-like activity originates in multiple areas of the ca2-ca3 region of hippocampal slices. J Neurophysiol 71: 1574–1585, 1994.
Demir R, Haberly LB, and Jackson MB. Voltage imaging of epileptiform activity in slices from rat piriform cortex: onset and propagation. J Neurophysiol 80: 2727–2742, 1998.
Demir R, Haberly LB, and Jackson MB. Characteristics of plateau activity during the latent period prior to epileptiform discharges in slices from rat piriform cortex. J Neurophysiol 83: 1088–1098, 2000.
Durand GM, Kovalchuk Y, and Konnerth A. Long-term potentiation and functional synapse induction in developing hippocampus. Nature 381: 71–75, 1996.[CrossRef][Medline]
Gaiarsa JL, Corradetti R, Cherubini E, and Ben-Ari Y. Modulation of GABA-mediated synaptic potentials by glutamatergic agonists in neonatal CA3 hippocampal neurons. Eur J Neurosci 3: 301–309, 1991.[CrossRef][Web of Science][Medline]
Ganguly K, Schinder AF, Wong ST, and Poo M. GABA itself promotes the developmental switch of neuronal GABAergic responses from excitation to inhibition. Cell 105: 521–532, 2001.[CrossRef][Web of Science][Medline]
Garaschuk O, Hanse E, and Konnerth A. Developmental profile and synaptic origin of early network oscillations in the CA1 region of rat neonatal hippocampus. J Physiol 507: 219–236, 1998.
Garaschuk O, Linn J, Eilers J, and Konnerth A. Large-scale oscillatory calcium waves in the immature cortex. Nat Neurosci 3: 452–459, 2000.[CrossRef][Web of Science][Medline]
Gasparini S, Saviane C, Voronin LL, and Cherubini E. Silent synapses in the developing hippocampus: lack of functional AMPA receptors or low probability of glutamate release? Proc Natl Acad Sci USA 97: 9741–9746, 2000.
Goodman CS and Shatz CJ. Developmental mechanisms that generate precise patterns of neuronal connectivity. Cell 72: 77–98, 1993.[CrossRef][Web of Science][Medline]
Grinvald A, Manker A, and Segal M. Visualization of the spread of electrical-activity in rat hippocampal slices by voltage-sensitive optical probes. J Physiol 333: 269–291, 1982.
Huang X, Troy WC, Yang Q, Ma H, Laing CR, Schiff SJ, and Wu JY. Spiral waves in disinhibited mammalian neocortex. J Neurosci 24: 9897–9902, 2004.
Iijima T, Witter MP, Ichikawa M, Tominaga T, Kajiwara R, and Matsumoto G. Entorhinal-hippocampal interactions revealed by real-time imaging. Science 272: 1176–1179, 1996.[Abstract]
Jin WJ, Zhang R, and Wu JY. Voltage-sensitive dye imaging of population neuronal activity in cortical tissue. J Neurosci Methods 115: 13–27, 2002.[CrossRef][Web of Science][Medline]
Kasyanov AM, Safiulina VF, Voronin LL, and Cherubini E. GABA-mediated giant depolarizing potentials as coincident detectors for enhancing synaptic efficacy in the developing hippocampus. Proc Natl Acad Sci USA 101: 3967–3972, 2004.
Katz LC and Shatz CJ. Synaptic activity and the construction of cortical circuits. Science 274: 1133–1138, 1996.
Khalilov I, Dzhala V, Ben-Ari Y, and Khazipov R. Dual role of GABA in the neonatal rat hippocampus. Dev Neurosci 21: 310–319, 1999.[CrossRef][Web of Science][Medline]
Khalilov I, Esclapez M, Nedina I, Aggoun D, Lamsa K, Leinekugel X, Khazipov R, and Ben-Ari Y. A novel in vitro preparation: the intact hippocampal formation. Neuron 19: 743–749, 1997.[CrossRef][Web of Science][Medline]
Khazipov R, Khalilov I, Tyzio R, Morozova E, Ben-Ari Y, and Holmes GL. Developmental changes in GABAergic actions and seizure susceptibility in the rat hippocampus. Eur J Neurosci 19: 590–600, 2004.[CrossRef][Web of Science][Medline]
Khazipov R, Leinekugel X, Khalilov I, Gaiarsa JL, and Ben-Ari Y. Synchronization of GABAergic interneuronal network in CA3 subfield of neonatal rat hippocampal slices. J Physiol 498: 763–772, 1997.
Kosmidis E, Cohen LB, Falk C, Wu JY, and Baker B. Imaging with voltage-sensitive dyes: spike signals, population signals, and retrograde transport. In: Imaging Living Cells, edited by Yuste R, Lanni F, and Konnerth A. New York: Cold Spring Harbor Press, 2004, p. 445–455.
Lauri SE, Lamsa K, Pavlov I, Riekki R, Johnson BE, Molnar E, Rauvala H, and Taira T. Activity blockade increases the number of functional synapses in the hippocampus of newborn rats. Mol Cell Neurosci 22: 107–117, 2003.[CrossRef][Web of Science][Medline]
Leinekugel X, Khalilov I, Ben-Ari Y, and Khazipov R. Giant depolarizing potentials: the septal pole of the hippocampus paces the activity of the developing intact septohippocampal complex in vitro. J Neurosci 18: 6349–6357, 1998.
Leinekugel X, Khazipov R, Cannon R, Hirase H, Ben-Ari Y, and Buzsaki G. Correlated bursts of activity in the neonatal hippocampus in vivo. Science 296: 2049–2052, 2002.
Leinekugel X, Medina I, Khalilov I, Ben-Ari Y, and Khazipov R. Ca2+ oscillations mediated by the synergistic excitatory actions of GABA(A) and NMDA receptors in the neonatal hippocampus. Neuron 18: 243–255, 1997.[CrossRef][Web of Science][Medline]
Leinekugel X, Tseeb V, Ben-Ari Y, and Bregestovski P. Synaptic GABAA activation induces Ca2+ rise in pyramidal cells and interneurons from rat neonatal hippocampal slices. J Physiol 487: 319–329, 1995.
Lippe WR. Rhythmic spontaneous activity in the developing avian auditory system. J Neurosci 14: 1486–1495, 1994.[Abstract]
Ma HT, Wu CH, and Wu JY. The initiation of spontaneous epileptiform events in the rat neocortex, in vivo. J Neurophysiol 91: 934–945, 2004.
Maffei L and Galli-Resta L. Correlation in the discharges of neighboring rat retinal ganglion cells during prenatal life. Proc Natl Acad Sci USA 87: 2861–2864, 1990.
McCormick DA. Brain calculus: neural integration and persistent activity. Nat Neurosci 4: 113–114, 2001.[CrossRef][Web of Science][Medline]
McCormick DA and Contreras D. On the cellular and network bases of epileptic seizures. Annu Rev Physiol 63: 815–846, 2001.[CrossRef][Web of Science][Medline]
Meister M, Wong RO, Baylor DA, and Shatz CJ. Synchronous bursts of action potentials in ganglion cells of the developing mammalian retina. Science 252: 939–943, 1991.
Menendez de la Prida L, Bolea S, and Sanchez-Andres JV. Analytical characterization of spontaneous activity evolution during hippocampal development in the rabbit. Neurosci Lett 218: 185–187, 1996.[CrossRef][Web of Science][Medline]
Menendez de la Prida L, Bolea S, and Sanchez-Andres JV. Origin of the synchronized network activity in the rabbit developing hippocampus. Eur J Neurosci 10: 899–906, 1998.[CrossRef][Web of Science][Medline]
Menendez de la Prida L and Sanchez-Andres JV. Non-linear frequency-dependent synchronization in the developing hippocampus. J Neurophysiol 82: 202–208, 1999.
Menendez de la Prida L and Sanchez-Andres JV. Heterogeneous populations of cells mediate spontaneous synchronous bursting in the developing hippocampus through a frequency-dependent mechanism. Neuroscience 97: 227–241, 2000.[CrossRef][Web of Science][Medline]
Miyakawa N, Yazawa I, Sasaki S, Momose-Sato Y, and Sato K. Optical analysis of acute spontaneous epileptiform discharges in the in vivo rat cerebral cortex. Neuroimage 18: 622–632, 2003.[CrossRef][Web of Science][Medline]
Momose-Sato Y, Mochida H, Sasaki S, and Sato K. Depolarization waves in the embryonic CNS triggered by multiple sensory inputs and spontaneous activity: optical imaging with a voltage-sensitive dye. Neuroscience 116: 407–423, 2003.[CrossRef][Web of Science][Medline]
O'Donovan MJ, Ho S, and Yee W. Calcium imaging of rhythmic network activity in the developing spinal cord of the chick embryo. J Neurosci 11: 6354–6369, 1994.
Owens DF, Boyce LH, Davis MB, and Kriegstein AR. Excitatory GABA responses in embryonic and neonatal cortical slices demonstrated by gramicidin perforated-patch recordings and calcium imaging. J Neurosci 16: 6414–6423, 1996.
Peinado A. Traveling slow waves of neural activity: a novel form of network activity in developing neocortex. J Neurosci 20: RC54, 2000.
Penn AA, Riquelme PA, Feller MB, and Shatz CJ. Competition in retinogeniculate patterning driven by spontaneous activity. Science 279: 2108–2112, 1998.
Serafini R, Valeyev AY, Barker JL, and Poulter MO. Depolarizing GABA-activated Cl channels in embryonic rat spinal and olfactory bulb cells. J Physiol 488: 371–386, 1995.
Spitzer NC and Gu X. Purposeful patterns of spontaneous calcium transients in embryonic spinal neurons. Semin Cell Dev Biol 8: 13–19, 1997.[CrossRef][Web of Science][Medline]
Strata F, Atzori M, Molnar M, Ugolini G, Tempia F, and Cherubini E. A pacemarker current in dye-coupled hilar interneurons contributes to the generation of giant GABAergic potentials in developing hippocampus. J Neurosci 17: 1435–1446, 1997.
Tominaga T, Tominaga Y, Yamada H, Matsumoto G, and Ichikawa M. Quantification of optical signals with electrophysiological signals in neural activities of Di-4-ANEPPS stained rat hippocampal slices. J Neurosci Methods 102: 11–23, 2000.[CrossRef][Web of Science][Medline]
Tsau Y, Guan L, and Wu JY. Initiation of spontaneous epileptiform activity in the neocortical slice. J Neurophysiol 80: 978–982, 1998.
Tsau Y, Guan L, and Wu JY. Epileptiform activity can be initiated in various neocortical layers: an optical imaging study. J Neurophysiol 82: 1965–1973, 1999.
Wong ROL. Retinal waves and visual system development. Annu Rev Neurosci 22: 29–47, 1999.[CrossRef][Web of Science][Medline]
Wu JY and Cohen LB. Fast multisite optical measurement of membrane potential. In: Biological Techniques: Fluorescent and Luminescent Probes for Biological Activity, edited by Mason WT. New York: Academic Press, 1993, p. 389–404.
Wu JY, Cohen LB, and Falk CX. Fast multisite optical measurement of membrane potential with two examples. In: Fluorescent and Luminescent Probes for Biological Activity, edited by Mason WT. New York: Academic Press, 1999a, p. 222–238.
Wu JY, Guan L, Bai L, and Yang Q. Spatiotemporal properties of an evoked population activity in rat sensory cortical slices. J Neurophysiol 86: 2461–2474, 2001.
Wu JY, Guan L, and Tsau Y. Propagating activation during oscillations and evoked responses in neocortical slices. J Neurosci 19: 5005–5015, 1999b.
Yuste R, Peinado A, and Katz LC. Neuronal domains in developing neocortex. Science 257: 665–669, 1992.
This article has been cited by other articles:
|
|
 |
|
  S. Jacobi and E. Moses Variability and Corresponding Amplitude-Velocity Relation of Activity Propagating in One-Dimensional Neural Cultures J Neurophysiol, May 1, 2007; 97(5): 3597 - 3606. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
