help button home button Endocrine Society Endocrinology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Endocrinology, doi:10.1210/en.2008-0299
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Haneda, K.
Right arrow Articles by Oka, Y.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Haneda, K.
Right arrow Articles by Oka, Y.
Endocrinology Vol. 149, No. 7 3540-3548
Copyright © 2008 by The Endocrine Society

Coordinated Synchronization in the Electrically Coupled Network of Terminal Nerve Gonadotropin-Releasing Hormone Neurons as Demonstrated by Double Patch-Clamp Study

Kosuke Haneda and Yoshitaka Oka

Laboratory of Biological Signaling, Department of Biological Sciences, Graduate School of Science, the University of Tokyo, Tokyo 113-0033, Japan

Address all correspondence and requests for reprints to: Yoshitaka Oka, Department of Biological Sciences, Graduate School of Science, The University of Tokyo, Tokyo 113-0033, Japan. E-mail: okay{at}biol.s.u-tokyo.ac.jp.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
The peptidergic neurons play important roles such as neuromodulatory and neuroendocrine functions in the central nervous system. However, our knowledge about the organization and the function of the peptidergic neuromodulator systems is still very poor. The terminal nerve GnRH peptidergic neurons of a teleost, the dwarf gourami (Colisa lalia), serve as an excellent model system for such study. The cell bodies are large and make up a tight cell cluster, and the easy access to the cell bodies on the ventral surface of the brain makes the electrophysiological measurements in a precisely controlled manner. Here we show direct evidence to demonstrate the electrical coupling and the synchronization of the neural firing activity among the terminal nerve GnRH neurons by using the double patch-clamp recording technique. The electrical coupling coefficient was strong enough (ranged from 0.083 to 0.370) to synchronize spontaneous firings of GnRH neurons in the cluster. A model, in which the firings in the cluster occur within a small time window (dozens of milliseconds), was verified by using the serial loose-seal extracellular patch-clamp recordings and the cross-correlogram analysis. The present findings provide several insights for understanding the physiological mechanisms and functional significance of synchronized activities in the peptidergic and/or aminergic neuromodulator system as well as in the peptidergic neuroendocrine cells.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
NEUROMODULATION, WHICH usually brings about the modulation of ion channel functions and synaptic transmissions, occurs through the activation of metabolic pathways of the neurons and is considered the important physiological functions of peptidergic and aminergic neurons. Such neuromodulator systems are characterized by several features in common (1); clustering of cell bodies, spontaneous and regular pacemaker-like firings, and wide projection of axons in the brain. These characteristic features of modulator neurons have been intensively studied by using the terminal nerve (TN) GnRH peptidergic neurons of a teleost brain as a model system, which has conspicuous morphological features advantageous for the physiological analysis of single peptidergic neuron (2, 3, 4). It has been shown that the TN-GnRH neurons of a teleost, the dwarf gourami, make a tight cluster of neurons, each of which exhibits regular spontaneous action potentials whose frequency is rather constant among the different neurons in the same cluster (1). Furthermore, it has been suggested that this pacemaker frequency changes according to the physiological condition of the animal, e.g. arousal, motivational status, hormonal milieu, etc., and the firing frequency of spontaneous activity is reported to affect the efficacy of exocytotic GnRH release from GnRH neurons (5).

Although the neurophysiological mechanisms of generation and frequency modulation of pacemaker activities by autocrine/paracrine mechanisms have been intensively studied (2, 3, 4, 6), the physiological significance of the commonly expressed characteristic features mentioned above has not been studied and remains rather enigmatic. Because these features are shared by peptidergic neurons other than GnRH neurons as well as the aminergic neurons, all of which are thought to function as neuromodulators, it should be essential to study the physiological significance of these features to understand the neuromodulator functions. In the present study, we used the TN-GnRH neurons of the dwarf gourami as a simple model system, which enabled the precise investigation of both the anatomical and physiological interactions between the pair of neurons in the cluster. By using the double patch-clamp recording techniques (both whole-cell voltage/current clamp and loose-seal configurations), we have shown here both physiologically and morphologically that TN-GnRH neurons are electrically coupled, and the coupling is strong enough to synchronize firings in the cluster. The dye-coupling experiments using neurobiotin intracellular staining suggested the importance of the contacts among the dendritic processes as the basis for the electrical coupling. The electrical coupling via the neural processes may also underlie synchronized activities in the peptidergic neuroendocrine cells, which are considered essential for the control of neuroendocrine functions such as reproduction.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Preparation of acute brain block with exposed TN-GnRH neurons
Adult male and female dwarf gouramis (Colisa lalia), approximately 4 cm in standard length, were purchased from a local dealer. Each fish tank containing up to 20 fish was maintained at 27 C and 12-h light, 12-h dark cycle. They were anesthetized in MS 222 (3-aminobenzoic acid ethyl ester) and were quickly killed by decapitation. The animals were maintained and used in accordance with the guidelines of the Physiological Society of Japan and the University of Tokyo for the Use and Care of Experimental Animals. After the ventral meningeal membrane of the forebrain was carefully removed, thick brain block of about 500 µm containing TN-GnRH neurons were manually cut out with a razor blade in a low-Na Ringer solution consisting of (in millimoles): NaCl 40, KCl 4.0, MgCl2 8.5, MgSO4 1.0, CaCl2 0.17, NaHCO3 26, NaH2PO4 1.0, glucose 10, and sucrose 210. The blocks containing TN-GnRH neurons were allowed to recover for at least 1 h before the experiments in the standard external solution (in mM): 150 NaCl, 5.0 KCl, 1.3 MgSO4, 2.4 CaCl2, 10 HEPES, 10 glucose, (adjusted to pH 7.4 with NaOH). All the experiments were performed at room temperature (22–24 C).

In the present study, the TN-GnRH neurons completely exposed to the surface and could be clearly identified by their size (about 20–50 µm, compared with the surrounding cells about 10 µm, in diameter; see Ref. 6 for the visual identification of the TN-GnRH neurons) were electrophysiologically analyzed. The number of cells visible from the surface of the present experimental preparations were 6.9 ± 1.6 (n = 41, range 5–11). In the Nissl-stain preparations, however, the average number of TN-GnRH neurons is reported to be 10.4 ± 1.8 (n = 16, range 6–13) (7). This minor discrepancy could be attributable to either the inevitable loss of neurons during the preparations or the invisible neurons buried underneath the ones exposed to the surface.

Electrophysiological recording
The electrical recordings were performed with MultiClamp 700A amplifier (Molecular Devices, Foster City, CA). The detailed procedure of electrophysiological recordings has been described in the previous work (6). Whole-cell recordings were made using patch pipettes of 4–7 M{Omega} resistance containing (in mM): 145 K-gluconate, 0.3 EGTA, 10 HEPES, 1.3 MgATP, and 10 glucose (adjusted to pH 7.2 with KOH). Loose-seal extracellular recordings (8, 9) were made using the pipette of 0.6–3 M{Omega} resistance containing standard external solution. Current-clamp recordings were filtered at 2.0 kHz by low-pass Bessel filter and sampled at 10 kHz, whereas the voltage-clamp recordings were filtered at 1.0 kHz and sampled at 5.0 kHz using the Digidata 1322A and pCLAMP9.2 software (Molecular Devices).

The membrane resistance and capacitance in the whole-cell recordings were 100.0 ± 40.9 M{Omega} (n = 74) and 166.3 ± 57.3 pF (n = 46), respectively (mean ± SD). The agar bridge was used for the reference electrode, and the liquid junction potential of the gluconate-based pipette solution was 7.60 ± 0.22 mV (mean ± SD n = 6). However, because this junction potential counteracted the voltage error due to the series resistance, we neglected these voltage artifacts and did not carry out any corrections.

Intracellular staining
Neurobiotin (20 mM) was injected to the cell under the whole-cell patch clamp mode by passing 1nA short depolarizing pulses of 150 msec duration for 10 min at room temperature (10). After 1 h of incubation at 30 C, the slices were fixed overnight at 4 C in a fixative mixture of 4% paraformaldehyde, 0.2% picric acid, and 0.1% glutaraldehyde in 0.1M phosphate buffer (PB). Each brain block was treated as follows: 1) rinsed with 0.1 M PB containing 0.75% NaCl and 0.5% Triton X-100 (PBST) and then treated with 1% H2O2 for 30 min and with 1% sodium borohydride for 30 min; 2) rinsed with 0.1 M PBST and then incubated for 4 h in biotinylated horseradish peroxidase conjugated with avidin (Vectastain ABC elite; Vector Laboratories, Burlingame, CA) in 0.1 M PBST; 3) rinsed with 0.1 M PBST, and incubated in 200 µl of 0.1 M PBST containing 0.05% 3,3'-diaminobenzidine (Sigma, St. Louis, MO) for 30 min and then developed by adding 2 µl of 1% H2O2; and 5) rinsed three times with 0.1 M PB and observed under bright-field illumination.

Data analysis
Temporal relationship of the activities between a pair of neurons was analyzed using the following procedure. First, the membrane potential of each neuron was simultaneously sampled at 10 kHz, and the action potentials were detected by threshold search program in pClamp software (Molecular Devices). Second, by using a custom program written in C language, a cross-correlation histogram, or cross-correlogram, was calculated for 150 sec duration with a 1 msec bin width. Third, to compare the degree of correlation in different pairs, each cross-correlogram was normalized to the firing frequencies of both neurons by dividing the counts of each bin by the expected number of spikes assuming the independent firing of the pair of neurons (11, 12). The expected number of spikes, N, was calculated as N = {lambda}1 x {lambda}2 x {Delta}t x T, where {lambda}1 and {lambda}2 are the firing frequency (in Hertz) of cell 1 and cell 2, respectively, {Delta}t the bin width (1 msec here), and T the data duration of 150 sec. The value greater than 1 + 3 x SD indicates a significant positive correlation, where SD is the SD of the cross-correlogram of the spike trains (13).


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Coordinated neural activities among the TN-GnRH neurons
As already reported in previous studies (reviewed in Refs. 2, 3, 4), the characteristic spontaneous and regular pacemaker activities were observed in the whole-cell current clamp as well as the loose-seal patch extracellular recordings. The average firing frequencies during 30 sec ranged from 0.8 to 10.3 Hz (3.8 ± 2.0 Hz, mean ± SD, n = 347, from whole-cell and extracellular recordings). We used the loose-seal patch extracellular recordings because it enables the stable and less invasive recordings of the action currents of all the visible neurons in the cluster at the same time in a relatively short period. By using this method, it was shown that neurons in the same cluster had very similar firing frequencies (Fig. 1AGo; see also Fig. 2AGo), thus confirming our previous findings (cf. Fig. 4BGo of Ref. 1). Next, the coordination of firing frequencies was further assessed with the simultaneous recordings from two neurons. Figure 1BGo shows that the firing frequencies of the simultaneously recorded pair of neurons located within the same neuronal cluster were virtually identical. Furthermore, depolarization of the pair of neurons with high K+ perfusing solution increased the firing frequencies and keeping the firing frequencies equal (Fig. 1CGo). These observations strongly suggest the presence of mechanisms that synchronize the pacemaker firing frequencies among the neurons in the cluster.


Figure 1
View larger version (10K):
[in this window]
[in a new window]

  		FIG. 1. Average firing frequencies of TN-GnRH neurons. A, Firing frequencies of all neurons from the same cluster (groups 1–5) are summarized. Each recording was conducted with the loose-seal extracellular recording. The average frequency of each group is shown by a horizontal bar. B, Firing frequencies simultaneously recorded from 176 pairs of neurons are summarized. The faster frequency is shown on the horizontal axis. C, High K+ stimulations (10 and 20 mM) increased the firing frequencies of neuron pairs, keeping the averaged firing frequencies of the pair of neurons quite similar.

 

Figure 2
View larger version (8K):
[in this window]
[in a new window]

  		FIG. 2. Synchronized firings of TN-GnRH neurons. A and B, Loose-seal extracellular recording (A) or whole-cell recording (B) was conducted from a pair of neurons in a single cluster. The two neurons fired almost in phase. C, Spikelet observed when the action potential of one neuron (cell 2) did not coincide with that of the other (cell 1). The area surrounded by the rectangle in B is enlarged in both time and voltage scales. Note the 10-fold difference in voltage scales for cell 1 and cell 2. The vertical broken line indicates the peak of the action potential in cell 2.

 

Figure 4
View larger version (8K):
[in this window]
[in a new window]

  		FIG. 4. The electrical coupling revealed by double whole-cell patch clamp recordings. A, In voltage clamp experiments, voltage steps are applied in cell 1, and the current responses of the both neurons were indicated. Small number in traces indicate the sequence of traces. B, The current responses of one neuron (cell 2) were plotted against the voltage steps applied to another neuron (cell 1). The passive currents flowed bidirectionally (in both depolarizing and hyperpolarizing directions), and the amplitude and shape of the current responses were symmetrical in both directions (see cell 2). The open and filled circles indicate the passive currents flowing from cell 2 to cell 1 and from cell 1 to cell 2, respectively. C, The coupling conductance was defined as the slope of the graph of B, and it showed a wide range of variation. It should be noted that the conductances determined in the pair of neurons were almost identical (n = 30).

 
To study the synchronization of pacemaker activities more precisely, the trajectories of membrane potentials of the pair of neurons were compared. It is noteworthy that the action potentials in one neuron were likely to coincide with that of the other (typical examples are shown in Fig. 2Go, A and B). In 39 pairs of neurons, 79.5 ± 23.0% (mean ± SD) of the action potentials occurred simultaneously [calculated as the ratio of the number of correlated spikes of cross-correlogram (see Materials and Methods and below) to the number of total firings of the faster neuron]. When the action potentials failed to coincide, a small elevation of membrane potential (2–5 mV amplitude ripple or spikelet) was evoked nearly simultaneously with the firing of the companion neuron (see the traces in rectangle in Fig. 2BGo, which is enlarged in Fig. 2CGo). Here it should be noted that the peak of the spikelet always occurred with a small delay from the peak of the action potential in the companion neuron (Fig. 2BGo). This strongly suggests that the spikelets originate from synaptic connections and the extremely short synaptic delay (less than 1 ms) further suggest the occurrence of the electrical synapse.

The temporal relationships of the firing activities between the pairs were further quantitatively analyzed using the cross-correlogram analysis (see Materials and Methods). Eighty-two percent (121 of 147) of the cross-correlograms during 150 sec recordings had a single clear peak, indicating a remarkable tendency for synchronization (Fig. 3AGo). The remaining 8% (11 of 147) had two attenuated peaks, and 10% (15 of 147) had no significant peak. In the cross-correlogram with a single clear peak (Fig. 3AGo), it should be noted that the peak of the cross-correlogram was often observed with some deviation from the origin (zero delay). Immediately after loose-seal extracellular recording from the pair of neurons in Fig. 3AGo, the recording electrodes were changed between the two. This caused a mirror-image shift of the peak of the cross-correlogram (Fig. 3BGo, compare bars with interrupted and continuous lines), suggesting that there is a temporal relationship between the firing of the pair of neurons, i.e. there is a leader-follower relationship between the recorded pair of neurons. In the successive recordings from different pairs of neurons by using the loose-seal extracellular recordings, exchange of the recording electrodes made a similar mirror-image shift in the peak of the cross-correlogram. The firings of a pair of neurons were not completely synchronized but always occurred with a small delay. In addition, the occurrence of a single sharp peak of cross-correlogram indicates the stability of the sequence of firings (see below for further discussion).


Figure 3
View larger version (9K):
[in this window]
[in a new window]

  		FIG. 3. The characteristics of the cross-correlation histogram. Eighty-two percent (121 of 147) showed a single clear peak (A). Eight percent (11 of 147) showed two small peaks, and 10% (15 of 147) did not show obvious correlation (not shown). The data obtained from the whole cell recordings and the extracellular recordings are pooled because both methods gave essentially identical results. B, In the pair with a single clear peak, the peak of the histogram causes a mirror-image shift when the recording electrodes are changed between the two. Using the loose-seal extracellular recordings, successive recordings were possible in the interval of 150 sec.

 
Electrical coupling in the neuronal cluster
Next, we investigated the mechanisms underlying the coordinated activity in the neuronal cluster. The lack of effect of the mixture of chemical synaptic blockers, 50 µM 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX), 125 µM D-2-amino-5-phosphonovalerate (AP-5), and 100 µM bicuculline, on the synchronization of the firing activities excluded the involvement of glutamate and {gamma}-aminobutyric acid (GABA) chemical synaptic transmissions in this coordination (data not shown). It should be noted, however, that the average firing frequencies were decreased to 46.0 ± 5.2% (n = 6) of the initial level in the blocker cocktail, although the peak amplitude of the cross-correlograms was not affected (not shown). The results suggest that the synaptic inputs from other neuronal systems affect the intrinsic frequency of spontaneous pacemaker activities, although the coordination of the firings was not affected by the external synaptic inputs. The coordinated firings are thought to arise from the functional interactions among the TN-GnRH neurons (see Discussion).

Voltage clamp experiments were done in the double whole-cell patch clamp configuration, and it was revealed that the TN-GnRH neurons were connected with each other by the electrical coupling as follows (Fig. 4Go). Bidirectional symmetrical currents were observed in the neurons when depolarizing or hyperpolarizing steps were applied to the companion neuron. The I-V curve was constructed from such experiments, and the transjunctional conductance was defined as the slope of the I-V relationship around 0 mV (Fig. 4BGo), which ranged between 0.8 and 4.0 nS with the average of 2.0 nS (Fig. 4CGo, n = 30). We did not find any sexual difference in the transjunctional conductance. The conductance in both directions between the pair showed nearly equivalent values, indicating that the electrical coupling is symmetrical, or nonrectifying. The gap-junction uncouplers, carbenoxolone, and 18β-glycyrrhetinic acid pharmacologically blocked the currents through this gap junctional conductance (Fig. 5AGo) in a concentration-dependent manner. After 60 min of blockade by 100 and 500 µM carbenoxolone, the coupling currents decreased to 20.0 ± 7.7% (mean ± SEM, n = 3) and 3.1 ± 1.7% of the initial levels, respectively (Fig. 5BGo). Recovery from the blockade was rather hard to obtain because of the long-lasting effect of these uncouplers. Therefore, it was difficult to examine the effect of these gap junction uncouplers on the coordinated activities.


Figure 5
View larger version (18K):
[in this window]
[in a new window]

  		FIG. 5. Effects of the gap-junction uncoupler on the coupling currents. A, Coupling currents were monitored by applying –100-mV step pulses every 2 min in the voltage clamp recordings. The blockers [carbenoxolone (CBX), 100 and 500 µM] were continuously applied, starting from 10 min after the coupling current became stable. All the points in the figure indicate mean ± SEM of the percentage to the initial level (n = 3 for control, n = 3 for CBX 100 µM, n = 3 for CBX 500 µM). B, Coupling current after 60 min blockade of the uncouplers are summarized. 18β-GA, 18β-Glycyrrhetinic acid.

 
In the current clamp experiments, the coupling coefficient (CC) was defined as the ratio of voltage response of the two neurons when depolarizing or hyperpolarizing currents were injected in one neuron (Fig. 6AGo) (12). Among the TN-GnRH neurons examined, the CC ranged between 0.083 and 0.370 with an average of 0.189 (Fig. 6BGo, calculated from 15 pairs), being significantly large, compared with the noradrenergic neurons in the locus coeruleus (average of 0.012) (14) and the dopaminergic neurons in the substantia nigra pars compacta (average of 0.023) (15).


Figure 6
View larger version (10K):
[in this window]
[in a new window]

  		FIG. 6. Summary of results of the current clamp experiments for the measurement of coupling coefficients. A, In the current clamp recording, constant currents (from –300 to +300 pA) were injected to one cell, whereas the membrane potentials of both cells were recorded in the presence of tetrodotoxin (1 µM). B, Summary of CC (CC1 and CC2). CC1 is defined as the amplitude of voltage response in cell 1 divided by the voltage response in cell 2, and vice versa (n = 15).

 
To search for the presence of morphological correlates of the electrophysiologically demonstrated electrical coupling via gap junctions, we examined the possible transfer of low-molecular-weight intracellular tracer, neurobiotin, through the gap junctions (16). The dye-coupling of the tracer in the neuronal cluster was observed in 30% of the slices (six of 20 slices). As shown in Fig. 7Go, multiple labeling of GnRH neurons in contact with the neurobiotin-injected neuron through neuronal processes was observed, which suggests that the gap junctions may be formed among the dendritic processes of the TN-GnRH neurons. These morphological observations confirm the electrophysiological results that the TN-GnRH neurons are electrically coupled with each other and further suggest that the coupling can occur among the neurons that are not in direct somatosomatic contacts, which agrees well with the absence of gap junctions in the somatosomatic contact areas at the electron-microscopic level (17).


Figure 7
View larger version (120K):
[in this window]
[in a new window]

  		FIG. 7. Dye coupling among the TN-GnRH neurons. A single neuron (filled arrow) was injected with neurobiotin. Two neurons (open arrows) were labeled because of the dye-coupling via gap junctions. It should be noted that the cell-cell contacts among the injected and noninjected but labeled neurons occur through their somatic or dendritic processes and not the cell bodies. Scale bar, 50 µm. TE, Telencephalon; OB, olfactory bulb.

 
The leader-follower relationship examined with the cross-correlogram analysis
As described in Fig. 3Go, the action potentials of one neuron (the follower) are usually preceded by that of the other neuron (the leader) with a certain amount of time delay (up to several tens of milliseconds). We therefore examined more precisely the leader-follower relationship among the cluster of TN-GnRH neurons by using the cross-correlogram analysis. By taking advantage of the loose-seal extracellular recording, the time delays of the neuron pairs were surveyed from as many neurons as possible in one cluster (Fig. 8Go). The results show that the action potentials of the neurons in the same cluster always occurred in rather small time window (0–44 msec). In seven clusters examined in the same way, the delay times were distributed within several tens of milliseconds. It should be stressed here that these time delays are considerably smaller than the interspike interval of the pacemaker firings in the TN-GnRH neurons (about 100-1000 msec). Because the firings are defined as the timing of the action potential peaks in the cross-correlogram analysis and the mean time delays of firings include the generation process of the action potentials, the time delays shown in Fig. 8Go are considerably larger than the pure synaptic delay of electrical synapses mentioned in Fig. 2BGo.


Figure 8
View larger version (15K):
[in this window]
[in a new window]

  		FIG. 8. Histogram showing the mean time delay of synchronization in the same neuron cluster. In the cross-correlograms with a single peak, the mean time delay was calculated by fitting the cross-correlograms to a Gaussian function. In the neurons cluster shown here, the cluster of TN-GnRH neurons are composed of nine neurons, and the mean time delay in the 33 pairs of neurons was calculated. Twenty-eight of 33 cross-correlograms were categorized into the one with a single peak. Although the distribution of mean time delays ranged widely from 0 to 40 msec, they were much smaller than the interspike interval of the spontaneous firings in the TN-GnRH neurons, which ranged from about 100 to 1000 msec.

 
Another important result of the cross-correlogram analysis is as follows. By definition, the occurrence of the single peak in the cross-correlogram indicates that the leader-follower relationship is highly stable during the entire recording (about 150 sec in the analysis above). In addition, each pair of neurons in the same neuronal cluster had a different delay of synchronization (Fig. 8Go). Then it is reasonable to consider that the coordination in the neuronal cluster is temporally structured in the millisecond scale. If we assume that the neurons sequentially fire in a certain order in the cluster, the amplitude of the time delay among the neuronal cluster should follow a simple arithmetic rule: when three neurons in a cluster fire in the order cell A->cell B->cell C, the mean time delay between cell A and cell C should be equal to the sum of those of the two pairs in between so that the time delay (A->C) approximately equal to delay (A->B) + delay (B->C). Actually, among the neurons in a certain clusters, we observed that the rule applies, although in some of the clusters it did not. Especially, in the recordings separated by long (several tens of minutes) intervals of recording, the simple arithmetic rule did not apply. A possible explanation for this deviation from the rule is the time course of changes in the temporal relationship of the firings during the entire recordings. Figure 9Go clearly shows that the peak of the cross-correlogram changes during the progress of the recordings for a long time, which suggests that the time delay or the leader-follower relationship are not rigidly fixed in the cluster but are temporally variable.


Figure 9
View larger version (11K):
[in this window]
[in a new window]

  		FIG. 9. Time course of changes in the peak of the cross-correlograms. A, Loose-seal double extracellular recording of 150 sec were conducted at time intervals of 10 min for each pair of recording. The displacements of the peak from zero in the cross-correlograms are plotted against the total time of recording. After 50 min, the second peak appeared. B, Representative cross-correlograms in A at times 0 (black solid line), 30 (dotted line), and 60 min (gray line).

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Coordinated activity of GnRH neurons and the role of gap junctional communications
The coordination of activity in the group of peptidergic neurons and neuroendocrine cells has attracted interests of many scientists and has been the focus of intensive studies. In the GnRH neuronal systems, the mechanisms underlying the pulsatile release of GnRH from the hypothalamic GnRH neurons have been studied by using the immortalized GT1–7 cell line as a model system for the GnRH neurons. In the GT1–7 cells, gap junctions have been suggested to be involved in the synchronized electrical burst activities and exocytotic release events (18, 19), and the presence of gap junctions has been examined by the dye-coupling and the immunological detection of gap junctional proteins (connexins) (20). However, the occurrence of electrical coupling and synchronization of activities has not yet been examined by electrophysiological methods. In the present study, we provide the first direct physiological and morphological evidences for the electrical coupling and synchronization of firing activities in the authentic GnRH-secreting neurons in the brain.

Furthermore, the results obtained in the present study highlight the significance of the dendritic processes among the TN-GnRH neurons in the coordination of the group of peptidergic neurons. This leads us to reason that, in the TN-GnRH neurons of the other species with slightly scattered distribution of cell bodies, it is possible that they are also electrically coupled with each other through the dendritic processes. Furthermore, such electrical coupling via the dendritic processes could contribute to the coordination of the GnRH neurons of the hypothalamic preoptic area, which are more or less scattered and do not make cell cluster in the mammalian brains. In fact, morphological evidence in favor of this possibility has been reported in the mammalian hypothalamic GnRH neurons, but the functional significance of this remains to be studied (21, 22).

Synchronization of activity in the neuronal cluster
The relationship of the synchronization of neuronal network activity and the synaptic connections is extremely complex (23). Saraga et al. (24) demonstrated in their theoretical studies that the synchronous activity in two-cell network connected by the dendro-dendritic gap junction is influenced not only by the firing frequency and the coupling intensity but also by the electrophysiological and morphological properties of the interconnecting dendrites (24). Especially, in the firing with frequency range of lower than 10 Hz and with strong electrical coupling, they have shown that the in-phase synchronization was more stable than antiphase synchronization. In the present study, we have shown that their theoretical prediction of stable in-phase synchronization applies to the realistic physiological situations, the TN-GnRH neurons, which show lower firing frequency (0.5–10 Hz) and strong electrical coupling (~2 nS).

Furthermore, we found that the mean time delays of synchronous firings varied among the pairs of neurons in the cluster (Fig. 8Go). These delays were generally much larger (up to 20 msec), compared with the synaptic delay expected for the electrical synapses [less than 1 msec; see review by Bennett (25)]. A possible explanation for this variability in the time delay is that there are several preferential pathways for the action potential propagation in the neuronal cluster so that the actual lengths of the transmission pathways are different among the recorded pairs. Furthermore, although the onset of the coupling currents under the voltage clamp was virtually the same as that of the voltage steps (Fig. 4CGo), the electrical influence of the coupling currents under the natural conditions, which depolarize the membrane potential strongly enough to evoke firing in the electrically coupled neurons, should take some time because of the electrotonic nature of the membrane. Multineuronal recordings in a millisecond time scale such as the high-performance imaging (26) may be one of the promising future directions toward the understanding of the mechanisms of the coordinated neural activities in the cluster of peptidergic neurons.

Effects of electrical coupling on the coordinated activity of GnRH neurons as a cluster
There may be two different mechanisms of realizing synchronization of neuronal activities in a cluster: 1) via intrinsic interconnections among the components or 2) by the common neural inputs that simultaneously affect the entire components. The results of the present study that the mixture of chemical synaptic blockers did not affect the synchronization of the firing activities provide clear evidence that neither glutamate/GABA synaptic transmissions nor external synaptic inputs are not involved in the synchronization. We conclude that the electrical coupling via gap junctions underlie the synchronization of TN-GnRH neurons in the cluster, although the overall frequency of the neurons as a cluster may be affected by synaptic inputs. As has been shown in the previous electrophysiological studies, the mechanism is known to work in the synchronization of the neurons in the inferior olive (27, 28), the suprachiasmatic nucleus (29) and the locus coeruleus (14), etc. Also, the mechanism has been considered to underlie the synchronization of oxytocin neurons in the supraoptic nucleus (30, 31). It should be noted that the gap junctions provide a major means by which the neurons are coordinated in most of the interconnected networks mentioned above.

The electrical coupling generally functions to assimilate the membrane potentials among the neurons, which effectively works to tune the phase of the membrane potential oscillations. In addition, unlike the chemical synapses, the transmission via the gap junctions is nonrectifying, and they transmit the signal in both directions (as shown in Fig. 4Go). These properties enable the gap junctions to function as the electrical synapses and efficiently synchronize the neuronal network. That is, the sequence of firings is not necessarily fixed in the recorded pairs in the cluster of TN-GnRH neurons; the pre- and postsynaptic neurons are interchangeable in the electrically coupled neurons in the cluster. The variability of the leader-follower relationship such as that shown in Fig. 9Go could be partly attributed to these bidirectional properties of the electrical coupling.

Gap junctions for the communication of cytoplasmic signaling molecules
There is another important aspect of the gap junctional communications to be considered in addition to the electrical coupling effects. The occurrence of dye-coupling among the neurons (Fig. 7Go) has been generally considered as an indication of the presence of gap junctions, which not only underlies the electrical coupling among the neurons but also allow the diffusion of small molecules such as the second messengers and the metabolites (reviewed in Ref. 32). Furthermore, in the secretory processes of the endocrine and exocrine cells, the diffusion of metabolites through gap junctions has been suggested to play a key role in signaling and synchronization (33). Therefore, in addition to the electrical coupling effects of gap junctional communications, we should also take into consideration the possible occurrence of metabolic coupling among the TN-GnRH neurons, and its possible physiological functions are interesting future topics for investigation.

Functional significance of coordinated synchronization of electrically coupled neurons for the neuromodulator systems
As mentioned above, the TN-GnRH neurons in the cluster form a functional assembly via electrical coupling and resultant synchronization of spontaneous regular firing activities. This property of the cluster is suggested to serve as a key to the understanding of the physiological functions of the neuromodulatory TN-GnRH system. Our previous study demonstrated that each TN-GnRH neuron projects widely throughout the brain but with a tendency to project more heavily to some areas, with considerable overlaps among the different neurons (1). Therefore, it is possible that the clustering of the neurons functions as an averaging device for the individual neuronal activities, which leads to the transmission of a coordinated uniform output, i.e. synchronized regular firing activity at certain frequency, throughout the projection areas covering wide areas of the brain from the olfactory bulb to the rostral spinal cord.

Concerning the other neuromodulator systems, there have been rather limited data on the cellular architecture and mechanisms of coordination of the activity. Interestingly, both the noradrenergic neurons of the locus coeruleus (14) and the dopaminergic neurons of the substantia nigra pars compacta (15) have been reported to be electrically coupled. However, synchronous firings have been observed only in the former but not the latter. These aminergic systems are composed of several hundreds to thousands of neurons, which makes it difficult to simply compare with the results of TN-GnRH system, which consist of only about 10 neurons in one cluster. Nevertheless, the present study has proved the technical advantages of the present experimental system: the TN-GnRH neurons consist of rather small number of neurons, most of which are easily accessible for precisely controlled electrophysiological methods. This has not been possible for the classical neuromodulator systems because of the technical difficulties.

The TN-GnRH neurons are controlled in an autocrine/paracrine manner by the GnRH peptide synthesized and released from the TN-GnRH neurons themselves, which results in the synchronized positive feedback facilitation of multiple GnRH neurons (3, 4, 34). Thus, the TN-GnRH neurons of the dwarf gourami form a functionally coordinated cluster via diffusional signal (auto/paracrine regulation by GnRH) as well as the electrotonic and metabolic coupling via the gap junctional communications.


   Acknowledgments
 
We thank Dr. Park Min-Kun for his advice and discussion and Ms. Miho Kyokuwa for her technical advice.


   Footnotes
 
This work was supported by Grants-in-Aid from the Ministry of Education, Culture, Sports, Science and Technology (MEXT) (No. 18021009) and the Japan Society for the Promotion of Science (JSPS) (No. 18370029) and the Program for Promotion of Basic Research Activities for Innovative Biosciences (PROBRAIN) of Japan (to Y.O.) and Research Fellowships of the JSPS for Young Scientists (to K.H.). This paper is based on the research performed as a partial fulfillment for the Ph.D. requirements at the University of Tokyo.

Disclosure Statement: The authors have nothing to disclose.

First Published Online April 10, 2008

Abbreviations: CC, Coupling coefficient; GABA, {gamma}-aminobutyric acid; PB, phosphate buffer; PBST, NaCl and Triton X-100; TN, terminal nerve.

Received March 3, 2008.

Accepted for publication March 31, 2008.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Oka Y, Matsushima T 1993 Gonadotropin-releasing hormone (GnRH)-immunoreactive terminal nerve cells have intrinsic rhythmicity and project widely in the brain. J Neurocsi 13:2161–2176[Abstract]
  2. Oka Y 1997 The gonadotropin-releasing hormone (GnRH) neuronal system of fish brain as a model system for the study of peptidergic neuromodulation. In: Parhar IS, Sakuma Y, eds. GnRH neurons: genes to behavior. Tokyo: Brain Shuppan Publishers; 245–276
  3. Oka Y 2002 Physiology and release activity of GnRH neurons. Prog Brain Res 141:259–281[Medline]
  4. Abe H, Oka Y 2007 Neuromodulatory functions of terminal nerve-GnRH neurons. In: Hara TJ, Zielenski B, ed. Fish physiology. Vol 25. Sensory systems neuroscience. Chapter 11. San Diego: Elsevier Inc.; 455–503
  5. Ishizaki M, Iigo M, Yamamoto N, Oka Y 2004 Different modes of gonadotropin-releasing hormone (GnRH) release from multiple GnRH systems as revealed by radioimmunoassay using brain slices of a teleost, the dwarf gourami (Colisa lalia). Endocrinology 145:2093–2103
  6. Haneda K, Oka Y 2004 Selective modulation of voltage-gated calcium channels in the terminal nerve gonadotropin-releasing hormone neurons of a teleost, the dwarf gourami (Colisa lalia). Endocrinology 145:4489–4499[Abstract/Free Full Text]
  7. Oka Y, Munro AD, Lam TJ 1986 Retinopetal projections from a subpopulation of ganglion cells of the nervus terminalis in the dwarf gourami (Colisa lalia). Brain Res 367:341–345[CrossRef][Medline]
  8. Nunemaker CS, DeFazio RA, Moenter SM 2003 A targeted extracellular approach for recording long-term firing patterns of excitable cells: a practical guide. Biol Proced Online 5:53–62[CrossRef][Medline]
  9. Perkins KL 2006 Cell-attached voltage-clamp and current-clamp recording and stimulation techniques in brain slices. J Neurosci Methods 154:1–18[CrossRef][Medline]
  10. Kita H, Armstrong W 1991 A biotin-containing compound N-(2-aminoethyl)biotinamide for intracellular labeling and neuronal tracing studies: comparison with biocytin. J Neurosci Methods 37:141–150[CrossRef][Medline]
  11. Aertsen AHJ, Gerstein GL 1985 Evaluation of neural connectivity: sensitivity of cross-correlation. Brain Res 340:341–354[CrossRef][Medline]
  12. Mann-Metzer P, Yarom Y 1999 Electrotonic coupling interacts with intrinsic properties to generate synchronized activity in cerebellar networks of inhibitory interneurons. J Neurosci 19:3298–3306[Abstract/Free Full Text]
  13. Hayar A, Shipley MT, Ennis M 2005 Olfactory bulb external tufted cells are synchronized by multiple intraglomerular mechanisms. J Neurosci 25:8197–8208[Abstract/Free Full Text]
  14. Christie MJ, Williams JT, North RA 1989 Electrical coupling synchronizes subthreshold activity in locus coeruleus neurons in vitro from neonatal rats. J Neurosci 9:3584–3589[Abstract]
  15. Vandecasteele M, Glowinski J, Vanance L 2005 Electrical synapses between dopaminergic neurons of the substantial nigra pars compacta. J Neurosci 25:291–298[Abstract/Free Full Text]
  16. Vaney DI 1991 Many diverse types of retinal neurons show tracer coupling when injected with biocytin or neurobiotin. Neurosci Lett 125:187–190[CrossRef][Medline]
  17. Oka Y, Ichikawa M 1991 Ultrastructure of the ganglion cels of the terminal nerve in the dwarf gourami (Colisa lalia). J Comp Neurol 304:161–171[CrossRef][Medline]
  18. Funabashi T, Suyama K, Uemura T, Hirose M, Hirahara F, Kimura F 2001 Immortalized gonadotropin-releasing hormone neurons (GT1–7 cells) exhibit synchronous bursts of action potentials. Neuroendocrinology 73:153–165
  19. Vazquez-Martinez R., Shorte SL, Boockfor FR, Frawley LS 2001 Synchronized exocytotic bursts from gonadotropin releasing hormone-expressing cells: dual control by intrinsic cellular pulsatility and gap junctional communication. Endocrinology 142:2095–2101[Abstract/Free Full Text]
  20. Matesic DF, Germak JA, Dupont E, Madhukar BV 1993 Immortalized hypothalamic luteinizing hormone-releasing hormone neurons express a connexin 26-like protein and display functional gap junction coupling assayed by fluorescence recovery after photobleaching. Neuroendocrinology 58:485–492[Medline]
  21. Silverman AJ, Witkin JW 1995 Synaptic interactions of luteinizing hormone releasing hormone (LHRL) neurons in the guinea pig preoptic area. J Histochem Cytochem 33:69–72
  22. Witkin J 1999 Synchronized neuronal networks: the GnRH system. Microsc Res Tech 44:11–18[CrossRef][Medline]
  23. Neltner L, Hansel D, Mato G, Meunier C 2000 Synchrony in heterogeneous networks of spiking neurons. Neura Comput 12:1607–1641[CrossRef]
  24. Saraga F, Ng L, Skinner FK 2006 Distal gap junctions and active dendrites can tune network dynamics. J Neurophysiol 95:1669–1682[Abstract/Free Full Text]
  25. Bennett MVL 1977 Electrical transmission: a functional analysis and comparison to chemical transmission. In: Kandel ER, ed. Handbook of physiology. Section 1: the nervous system. Vol I. Cellular biology of neurons. Baltimore: Williams & Wilkins; 357–416
  26. Ikegaya Y, Le Bon-Jego M, Yuste R 2005 Large-scale imaging of cortical network activity with calcium indicators. Neurosci Res 52:132–138[CrossRef][Medline]
  27. Long MA, Deans MR, Paul DL, Connors BW 2002 Rhythmicity without synchrony in the electrically uncoupled inferior olive. J Neurosci 22:10898–10905[Abstract/Free Full Text]
  28. Leznik E, Llinàs R 2005 Role of gap junctions in synchronized neuronal oscillations in the inferior olive. J Neurophysiol 94:2447–2456[Abstract/Free Full Text]
  29. Long MA, Jutras MJ, Connors BW, Burwell RD 2005 Electrical synapses coordinate activity in the suprachiasmatic nucleus. Nat Neurosci 8:61–66[CrossRef][Medline]
  30. Israel JM, Masson GL, Theodosis DT, Poulain DA 2003 Glutamatergic input governs periodicity and synchronization of bursting activity in oxytocin neurons in hypothalamic organotypic cultures. Eur J Neurosci 17:2619–2629[CrossRef][Medline]
  31. Moos F, Marganiec A, Fontanaud P, Guillou-Duvoid Alonse G 2004 Synchronization of oxytocin neurons in suckled rats: possible role of bilateral innervation of hypothalamic supraoptic nuclei by single medullary neurons. Eur J Neursci 20:66–78[CrossRef][Medline]
  32. Harris AL 2007 Connexin channel permeability to cytoplasmic molecules. Prog Biophys Mol Biol l94:120–143
  33. Michon L, Nlend Nlend R, Bavamian S, Bischoff L, Boucard N, Caille D, Cancela J, Charollais A, Charpantier E, Klee P, Peyrou M, Populaire C, Zulianello L, Meda P 2005 Involvement of gap junctional communication in secretion. Biochim Biophys Acta 1719:82–101[Medline]
  34. Abe H, Oka Y 2002 Mechanisms of the modulation of pacemaker activity by GnRH peptides in the terminal nerve-GnRH neurons. Zool Sci 19:111–128[CrossRef][Medline]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Haneda, K.
Right arrow Articles by Oka, Y.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Haneda, K.
Right arrow Articles by Oka, Y.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Endocrinology Endocrine Reviews J. Clin. End. & Metab.
Molecular Endocrinology Recent Prog. Horm. Res. All Endocrine Journals