| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Neurology and Neurosurgery, Montreal Neurological Institute and McGill University, Montreal, Quebec, Canada
Submitted 10 January 2007; accepted in final form 26 August 2007
 |
ABSTRACT |
|---|
|
|
|
INTRODUCTION |
|---|
|
). Since then, the M-current has been shown to be present in many other cell types including neurons in the CNS (reviewed by Brown 1988). It is now known that M-channels belong to the Kv7 (KCNQ) gene family, are mainly composed of Kv7.2 and Kv7.3 subunits in an hetero-multimetric complex (Kv7.2/Kv7.3; Wang et al. 1998) with additional contributions from Kv7.2 homomers and Kv7.5/Kv7.3 heteromers (Hadley et al. 2003; Shah et al. 2002), and are ubiquitous in the brain (Cooper et al. 2001).
The low-threshold (below –60 mV), slowly activating and deactivating, and noninactivating properties of the M-current are suggestive of a role in regulating neural excitability. Using bullfrog sympathetic neurons, Adams et al. (1982a,b) demonstrated that the M-current helps "clamp" the membrane potential near rest, due to its subthreshold activation and persistent nature. Suppression of the M-current by muscarinic agonists or by selective blockers of the M-channels, such as linopirdine [3,3-bis(4-pyridinylmethyl)-1-phenylindolin-2-one (DUP996)] (Aiken et al. 1995; Costa and Brown 1997; Lamas et al. 1997; Schnee and Brown 1998) and XE991 [10,10-bis(4-pyridinylmethyl)-9(10H)-anthracenone] (Zaczek et al. 1998) causes an increase in intrinsic excitability. Spike frequency adaptation was reduced in hippocampal pyramidal cells (Aiken et al. 1995; Cole and Nicoll 1983; Madison and Nicoll 1984; Peters et al. 2005), neocortical neurons (McCormick and Williamson 1989), and superior cervical ganglion cells (Wang et al. 1998; but see Miles et al. 2005; Romero et al. 2004) with M-current suppression. Reduced spike afterhyperpolarization (AHP) in hippocampal CA1 pyramidal neurons (Gu et al. 2005; Storm 1989), superior cervical ganglion (Wang et al. 1998), and rat ventral tegmental area dopamine neurons (Koyama and Appel 2006) has also been demonstrated.
Two lines of recent studies have shown that the M-current controls membrane dynamics of neurons in addition to its classical role in the control of excitability. Yue and Yaari (2004) have shown that the suppression of the M-current by linopirdine shifts the firing mode of the hippocampal CA1 pyramidal cells from regular firing to burst firing by augmenting the spike afterdepolarization (ADP). Hu et al. (2002) and Peters et al. (2005) have shown that suppression of the M-current in hippocampal pyramidal cells reduces intrinsic subthreshold theta resonance. Spike bursts and subthreshold membrane potential oscillations are believed to be important for synaptic plasticity (Magee and Johnston 1997; Thomas et al. 1998) and network oscillation (Buzsáki 2002; Fransén et al. 2004). These studies suggest the importance of the M-current not only in excitability control but also in brain functions such as memory. However, the role of the M-current in the entorhinal cortex (EC) has been investigated to a relatively lesser degree.
The EC, located in the temporal lobe between the hippocampus and the cortical mantle, is important for aspects of memory function (Leonard et al. 1995; Scoville and Milner 1957; Squire and Zola-Morgan 1991; Suzuki et al. 1997), including associative memory (Buckmaster et al. 2004) and spatial memory (Steffenach et al. 2005). Principal neurons in the EC receive cholinergic projections from the basal forebrain (Alonso et al. 1996) and cholinergic modulation is crucial for tuning the temporal lobe to mnemonic function (Hasselmo 1999). Being suppressed by cholinergic activation, the M-current could play an important role in the cholinergic regulation of mnemonic function. EC principal cells have layer- and cell-type–specific electrophysiological phenotypes (Alonso and Klink 1993; Dickson et al. 1997; Hamam et al. 2000), as well as dynamic properties such as subthreshold membrane potential oscillations (Alonso and Llinás 1989), delayed firing (Klink and Alonso 1997), and persistent firing (Egorov et al. 2002), all of which are demonstrated to be important for memory function (Fransén et al. 1999, 2002; McGaughy et al. 2005). Modulation of the dynamic properties of neurons through the M-current might thus be crucial in tuning the activity of the EC for memory function. Moreover, anatomical connections between the other cortical regions and the EC (Burwell and Amaral 1998a,b; Insausti et al. 1987, 1997) and between the EC and the hippocampus (Naber et al. 2001; Schwartz and Coleman 1981; Tamamaki and Nojyo 1993; Van Groen and Lopes da Silva 1986; van Groen et al. 1986; Witter and Amaral 1991) are restricted to specific layers of the EC. This suggests that layer-specific modulation might be important.
Using whole cell patch recording techniques in an EC slice preparation, we examined the effects of the M-current blocker linopirdine (10 µM) and XE991 (5 and 10 µM) on the firing behavior, intrinsic excitability, and subthreshold membrane potential oscillations of stellate cells (SCs) and nonstellate cells (non-SCs) from layer II, and of cells from layers III and V. We found that blockade of the M-current changed the firing pattern of the layer II non-SCs and layer III neurons from regular to intermittent firing. Whereas layer II non-SCs fired in high-frequency short-duration bursts nested in long-duration clustered firing, layer III cells fired regularly in a long-duration cluster during the intermittent firing mode. When three consecutive spikes were evoked at 50 Hz, a slow ADP developed in these two types of neurons resulting in delayed firing. As for the intrinsic excitability, spike frequency adaptation decreased significantly in layer III and layer V cells but not in layer II cells. Subthreshold membrane potential oscillations in the layer V cells were greatly suppressed, whereas those in layer II SCs remained intact. These results suggest that the M-current shapes the firing behavior, excitability, and subthreshold membrane potential oscillations in the EC in a cell-type–specific manner. Some of the results of this paper were previously reported in abstract form (Yoshida and Alonso 2005).
|
|
METHODS |
|---|
|
Experimental protocols were approved by the McGill University Animal Care Committee and the Institutional Animal Care and Use Committee at Boston University and were in compliance with guidelines of the Canadian Council on Animal Care. Long–Evans rats (postnatal days 21–27; Charles River, Quebec or Wilmington, MA) were anesthetized with ketamine/xylazine through intraperitoneal injection and intracardially perfused with ice-cold modified artificial cerebrospinal fluid (ACSF) containing (in mM) 110 choline chloride, 2.5 KCl, 1.25 NaH2PO4, 26 NaHCO3, 0.5 CaCl2, 7 MgCl2, 7 glucose, 3 pyruvic acid, and 1 ascorbic acid (pH adjusted to 7.4 by saturation with 95% O2-5% CO2). The brain was then removed from the cranium and placed in ice-cold modified ACSF. Horizontal slices (350 µm thick) of the hippocampal–entorhinal region were cut using a vibratome (Pelco series 1000, Leica VT 1000S, or World Precision Instruments Vibroslicer) and then transferred to a holding chamber, where they were kept submerged for >1 h at room temperature before recording. The holding chamber contained an extracellular solution containing (in mM) 124 NaCl, 3 KCl, 1.25 NaH2PO4, 26 NaHCO3, 1.6 CaCl2, 1.8 MgSO4, and 10 glucose (pH adjusted to 7.4 by saturation with 95% O2-5% CO2).
Recording
Whole cell recordings were obtained while slices were maintained in a submerged recording chamber at 30 ± 1°C and perfused with extracellular solution containing the glutamatergic and GABAergic ionotropic receptor blockers kynurenic acid (2 mM) and picrotoxin (100 µM), respectively. Stock solutions of linopirdine (10 mM, in ethanol), XE991 (5 mM, in water), and XE991 (10 mM, in water) and apamin (300 µM, in water) were prepared and diluted 1,000 times in the extracellular solution to make final concentrations of 10 µM, 5 µM, 10 µM, and 300 nM, respectively.
Patch pipettes were fabricated from borosilicate glass capillaries by means of a Sutter P-97 or P-87 horizontal puller. The patch pipettes were filled with intracellular solution containing (in mM) 120 K-gluconate, 10 HEPES, 0.2 EGTA, 20 KCl, 2 MgCl, 7 phosphocreatine-diTris, 4 Na2ATP, and 0.3 TrisGTP (pH adjusted to 7.3 with KOH). The intracellular solution also contained 0.1% biocytin for the purpose of labeling. When filled with this solution, the patch pipettes had a resistance of 4–8 M
. Slices were visualized with an upright microscope (Nikon E600FN), equipped with a x60 water-immersion objective lens, Nomarski optics, and a Newvicon camera (Dage-MTI NC-70), or with an upright microscope (Zeiss Axioskop 2), equipped with a x40 water-immersion objective lens, and a near-infrared charge-coupled device (CCD) camera (JAI CV-M50IR). Each layer was visually distinguished and the location of the cell was confirmed by biocytin staining after recording. Layer II SCs and non-SCs were distinguished by their morphology. Tight seals (>1 G) were formed on cell bodies and the membrane was ruptured with negative pressure. Current-clamp recordings were made with an Axopatch 1D, Axopatch 200B, or Multi Clamp 700B amplifier (Axon Instruments). Signals were filtered at 10 kHz and sampled at 20 kHz using Clampex 9.0 software (Axon Instruments). Drugs were purchased from Sigma and Tocris.
Data analysis
Clampfit 9.0 (Axon Instruments), IGOR Pro 5.0 (Wavemetrics), and Matlab (The MathWorks) were used for data analysis. Input resistance was measured from the voltage deflection in response to a 5-pA hyperpolarizing current pulse injection at a membrane potential of –60 mV. ADP amplitude was measured as the difference between the peak voltage of ADP and the resting potential. In cells with no clear peak of ADP (mainly layer III and layer V cells), peak voltage was measured at the mean time of the ADP peak of layer II cells (8 ms after the onset of current pulse). To measure the ADP duration and area, the onset of ADP was defined as the time of the fast AHP (indicated by an arrow in Fig. 2Aii), or the time of the largest deflection (indicated by an arrow in Fig. 2Cii) if the fast AHP was not evident. The duration of ADP was measured as the time from the onset of ADP to the time the membrane potential returned to the resting potential. The ADP area was measured as the integration of the difference between the membrane potential and the resting potential for the duration of ADP. The AHP amplitude was measured as the difference between the potential at negative peak of AHP and the resting potential. In cells where AHP was observed only in the control, which often happened in layer III and layer V cells, measurement at the time of peak in control was used as the AHP amplitude under M-current block. For each cell, these values were measured using the average of five traces in each condition.
|
|
Fig. 4, the afterpotential area was measured as the integration of the difference between the membrane potential and the resting potential (shown by dashed lines in Fig. 4ii) from the time of ADP onset to 30 s after current injection. The afterpotential areas measured in five traces were averaged for each cell.
|
|
Fig. 6, Biv–Bvi) were obtained from the same set of membrane potential traces used for the analysis of the intrinsic excitability. Bin size of the histogram was 20 ms. Histograms were normalized by the total number of ISIs.
|
|
). The membrane potential was kept at this level for 
20 s (54 s on average). Because spikes occasionally occurred at this membrane potential, we used a short (3.28 s) window of membrane potential that does not include spikes to obtain the power spectrum of the subthreshold membrane potential oscillation without interference from spikes. Our Matlab code obtained power spectra from multiple 3.28-s windows by sliding the window by 50-ms intervals throughout the recording period. Power spectra were obtained by fast Fourier transform using a Hanning window. Any window that contained spikes was eliminated. For each window, the area between 1.53 and 3.97 Hz in its power spectrum was calculated. Three nonoverlapping windows with the three largest areas were selected automatically. The frequency range (1.53–3.97 Hz) was chosen because the peak frequency of the subthreshold membrane potential oscillations fell into this range. Figure 7, i and ii, bottom traces and Fig. 7, iii and iv show, respectively, examples of windows and their power spectra. Three areas from three membrane potential windows were averaged for each condition, and average values and SE among the cells were plotted in the Fig. 7v.
|
P < 0.05, **0.001 P < 0.01, ***P < 0.001) was used. Data are expressed as means ± SE.
|
|
|
RESULTS |
|---|
|
In hippocampal CA1 neurons, blockade of the M-current by linopirdine switches their firing pattern from tonic to burst firing (Yue and Yaari 2004). We thus first explored how the M-current block affected the intrinsic firing pattern of the principal neurons in the EC. Principal cells were categorized into four groups throughout this study: 1) layer II SCs, 2) layer II non-SCs, 3) layer III pyramidal cells, and 4) layer V cells (including pyramidal, horizontal, and polymorphic cells). Layer V cells were not subdivided into three groups because there is no clear link between morphological characteristics and electrophysiological characteristics (Hamam et al. 2000). All the recordings were done in the presence of glutamatergic and GABAergic ionotropic receptor blockers. Input resistances of the four groups of cells were 119.4 ± 8.1, 146.0 ± 22.3, 477.2 ± 28.3, 435.5 ± 50.2 M, respectively.
The resting membrane potential of the four groups of cells varied from –64.8 to –61.9 mV under control conditions (Table 1) and no cell showed spontaneous firing. To examine their firing pattern under control conditions, cells were depolarized by constant-current injection to just above the threshold level (Fig. 1, Ai–Di). Minimum current that could cause at least five spikes in 15 s was chosen. When the cells responded with intermittent firing or bursts of spikes, this current caused at most 31 spikes in 15 s. In some cases, stronger depolarization was also induced (data not shown). After termination of the current injection, we bath-applied 10 µM linopirdine. Linopirdine blocks the Kv7 family (Brown et al. 2002), including the Kv7.2/Kv7.3 channel (Wang et al. 1998), from the extracellular side (Costa and Brown 1997; Lamas et al. 1997), and is 10-fold more selective for the M-current over other K+ channels, such as the delayed rectifier, transient, and Ca2+-activated K+ channels (Lamas et al. 1997; Schnee and Brown 1998). At this concentration, linopirdine blocks 60–90% of the M-current, depending on cell types (Aiken et al. 1995; Costa and Brown 1997; Lamas et al. 1997; Schnee and Brown 1998). The resting membrane potential slowly depolarized for 15–30 min, and the effect of the blockade was clear 30 min after application of linopirdine. The depolarization of the resting membrane potential often caused spontaneous firing. The membrane potentials in linopirdine were measured between firings of action potentials and varied from –57.5 to –50.6 mV (Table 1). To compare firing patterns among cell types and control conditions, cells were injected with a depolarizing or hyperpolarizing constant current to maintain the membrane potential just above the threshold level (Fig. 1, Aii–Dii) and at higher levels (data not shown).
|
). This mixed mode of subthreshold membrane potential oscillations and firing was observed up to –51.8 ± 1.4 mV (n = 6). Further depolarization caused tonic firing and subthreshold membrane potential oscillations no longer coexisted with spikes. In linopirdine, the subthreshold membrane potential oscillations and clustered firing remained intact in all eight cells at just above the threshold level (–53.9 ± 0.8 mV; Fig. 1Aii). The mixed mode was observed up to –52.6 ± 1.0 mV (n = 6). Further depolarization caused tonic firing and no cells switched to spike burst mode. In control conditions, all layer II non-SCs (n = 12) fired regularly, with no prominent subthreshold membrane potential oscillations, at just above the threshold level (–53.6 ± 0.8 mV; Fig. 1Bi) and at more depolarized levels (n = 10). In linopirdine, 75% (9 of 12) of layer II non-SCs switched to burst firing at just above the threshold level (–53.0 ± 0.8 mV; Fig. 1Bii). They tended to generate intermittent firing with long-duration (5.6 ± 0.6 s) clustered firing, which consisted of short-duration high-frequency bursts as shown in the inset in Fig. 1Bii. The high-frequency burst was most often a doublet but sometimes consisted of more than three spikes. The highest intraburst frequency was 45 ± 11 Hz. This nested bursting pattern was observed at membrane potentials up to –50.2 ± 2.1 mV (n = 4), and the high-frequency bursts alone (without intermittent firing) were observed with further depolarization. The remaining (3 of 12) layer II non-SCs fired regularly in linopirdine.
In control conditions, all layer III cells (n = 14) showed regular firing at just above the threshold level (–55.2 ± 0.7 mV; Fig. 1Ci) and at more depolarized levels (n = 8). In linopirdine, 86% (12 of 14) of these cells showed intermittent firing with long-duration (4.3 ± 1.3 s) low-frequency clustered firing at just above the threshold level (–54.9 ± 0.9 mV; Fig. 1Cii). The highest intracluster frequency was 9.2 ± 1.9 Hz. Depolarization during the clustered firing was larger than that in layer II non-SCs, resulting in a firing pattern similar to that of up and down states observed in slow-wave sleep (Steriade et al. 1993). In 33% (4 of 12) of the cells, this intermittency resulted in a sinusoidal oscillation. Whereas layer II non-SCs fired in high-frequency bursts, layer III cells did not show high-frequency bursts and fired regularly in the depolarized part of intermittent firing. The intermittent firing of these cells was observed at membrane potentials up to –51.2 ± 0.6 mV (n = 8) and showed tonic firing with further depolarization. The remaining layer III cells (2 of 14) fired regularly in linopirdine.
In control conditions, 53% (9 of 17) of the layer V cells had clear subthreshold membrane potential oscillations at just above the threshold level (–52.0 ± 0.8 mV). They often fired an action potential at the peaks of the subthreshold membrane potential oscillations (Fig. 1Di). This mixed mode of subthreshold membrane potential oscillations and spikes was observed up to –45.1 ± 1.0 mV (n = 5) and further depolarization caused tonic firing. The other cells (8 of 17) fired regularly without subthreshold membrane potential oscillations. In linopirdine, subthreshold membrane potential oscillations were greatly suppressed in all cells (n = 9); 88% (15 of 17) of the cells fired regularly (Fig. 1Dii) and the rest showed an intermittent firing pattern similar to that of the layer III cells at just above the threshold level (–51.9 ± 0.7 mV). However, none of them showed sinusoidal membrane oscillations.
In summary, the M-current block depolarized the membrane potential of all groups of cells. Depolarization was greater (>10 mV) in the layer III and layer V cells that had greater (>400 M) input resistances compared with the layer II cells that had much smaller (<150 M) input resistances, suggesting that input resistances of different groups of cells caused the different degrees of depolarization. The firing patterns of layer II non-SCs and layer III cells switched from regular to intermittent firing. In addition, whereas layer III cells fired regularly in the depolarized part of the intermittent firing, layer II non-SCs fired in high-frequency bursts. Finally, layer V cells showed a great suppression of subthreshold membrane potential oscillations. These data suggest that the modulation of the M-current shaped the intrinsic firing pattern of the EC principal neurons in multiple, possibly different ways.
To verify that these results are not due to rundown of the M-channels during whole cell recording, we performed control experiments in five cells in all cell groups. In the experiments with linopirdine (cited earlier), control recordings were started right after rupturing of the membrane and completed usually within 15 min. Linopirdine was applied right after the completion of the control recordings and recordings in linopirdine were started 30 min after the onset of the linopirdine application. Recordings in linopirdine were thus conducted 45 min after the rupturing of the membrane. To perform the control experiments without linopirdine with the same time course as in the linopirdine study, we conducted the first recordings right after the membrane rupturing (corresponding to the control recording) and the second recordings at 45 min after the rupturing of the membrane (corresponding to the recordings in linopirdine). We compared the firing pattern and membrane potential obtained from the first recordings (right after rupturing) and the second recordings (45 min after rupturing). In contrast to the preceding results in linopirdine, no cell group in these control experiments showed a significant change of membrane potential (data not shown) and no cell showed spontaneous firing. All cell types fired in the same manner as in the control and no cell switched to intermittent or burst firing.
Modulation of single-spike afterpotential
The M-current block enhances ADP (Yue and Yaari 2004) and decreases AHP (Gu et al. 2005; Koyama and Appel 2006; Storm 1989; Wang et al. 1998). Particularly in CA1 pyramidal cells, enhanced ADP reaches the firing threshold and leads to burst firing (Yue and Yaari 2004). We thus observed single-spike afterpotentials from each group of EC cells to investigate their contribution to the burst firing observed earlier. Short-duration (1-ms) depolarizing current pulses, with an amplitude just sufficient to elicit one action potential, were applied in control (gray traces in Fig. 2, Ai–Di) and in linopirdine (30 min; black traces in Fig. 2, Ai–Di). To quantitatively measure and compare ADP and AHP among the cell groups, all recordings were done at a membrane potential of –60 mV.
In layer II SCs (n = 7), responses from spikes were very similar in the control conditions and in linopirdine in 4-s traces (Fig. 2Ai). However, as shown in the magnified fast ADP part (Fig. 2Aii), ADP amplitude increased significantly with linopirdine (Fig. 2Aiii). In one of the seven cells, which showed a doublet in the previous section, the ADP reached spike threshold and one action potential was elicited on top of the ADP, even under control conditions. The ADP amplitude was not sufficient to reach spike threshold in linopirdine in the other six cells. The decay phase of the ADP in linopirdine was almost identical to, or in some cells even faster than, that in the control conditions (Fig. 2Aii). On average, neither the ADP duration nor the ADP area changed significantly with linopirdine (Fig. 2, Aiv and Av). The AHP amplitude also did not change significantly with linopirdine (Fig. 2Avi).
In layer II non-SCs (n = 9), the ADP amplitude in linopirdine also increased significantly (Fig. 2, Bi–Biii). Unlike in layer II SCs, repolarization after the peak of the ADP was usually slower in linopirdine than in control (Fig. 2, Bii and Biv). This caused an increase in the ADP area (Fig. 2Bv). The AHP amplitude decreased significantly with linopirdine (Fig. 2Bvi). In spite of what we expected, the increase in ADP amplitude was not sufficient to reach spike threshold, unlike the data for CA1 pyramidal cells (Yue and Yaari 2004).
The shape of the ADP of layer III cells was different from that of layer II cells (Fig. 2, Ci and Cii, n = 11). The ADP amplitude, measured at the average time of the layer II ADP peak, showed the largest increase among all cell types with linopirdine application (Fig. 2Ciii). Forty-five percent (5 of 11) of layer III cells did not have AHP, even in the control condition. Lack of an AHP resulted in the longest ADP duration and largest ADP area in control among all cell types (Fig. 2, Civ and Cv; note use of different scales compared with other cell types). Application of linopirdine further slowed the repolarization, resulting in an even larger ADP reaching 3 s in duration (Fig. 2, Ci and Civ). The increase in the ADP area (Fig. 2Cv; 991 mV·ms on average) was the largest among all four cell groups. With the application of linopirdine, the AHP became a large ADP in all cells (Fig. 2Cvi). The AHP amplitudes of the five cells with no AHP in control were measured at the average time of AHP in the other six cells.
Layer V cells showed significant increases in ADP amplitude, ADP duration, and ADP area (Fig. 2, Di–Dv). However, the ADP duration and ADP area were much smaller than those of layer III cells in linopirdine, being comparable to those of layer III cells in control conditions. AHP amplitude decreased significantly, often turning into an ADP (Fig. 2Dvi). The increase of ADP and decrease of AHP were not observed in the control study conducted without linopirdine (data not shown). In fact, some of the parameters showed average changes in control experiments that were in the opposite direction from that caused by linopirdine. These results indicate that the increase of ADP and decrease of AHP were not caused by rundown of the M-current.
What is the mechanism for the smaller AHP and larger ADP with blocking of the M-current? One possibility is that the lack of an M-current activated by an action potential reduced the repolarization drive, resulting in a larger ADP. However, the time constant for activation of the M-current is slow, ranging from 6 to 50 ms, depending on the membrane potential (Brown and Adams 1980; Wang et al. 1998), suggesting that the duration of an action potential is too short for full activation. To gain some insight into this, we elicited single spikes from –80 and –60 mV in layer V cells (n = 6). We found that the ADP did not increase in size with the application of linopirdine when the spike was elicited from –80 mV (Fig. 3A). ADP duration and ADP area did not change significantly at –80 mV (Fig. 3, B and C). Similar results were also observed in layer III cells (data not shown). If one action potential can activate the M-current, it should be possible to detect an increase in ADP with the application of linopirdine even at –80 mV because the shape of the action potential does not greatly vary at –80 and –60 mV. Therefore the lack of increase in ADP at –80 mV suggests that a single action potential was not able to activate the M-current. The ADP increase and AHP decrease observed at –60 mV are therefore likely attributable to a lack of M-current activation during the ADP as previously proposed (Yue and Yaari 2006) and not during the action potential.
To summarize, all cell types showed increases in fast ADP amplitudes and layer III cells showed a long-lasting, large-amplitude ADP in linopirdine. However, despite our above-mentioned expectation that enhanced ADP reaches the firing threshold and leads to burst firing, the ADP did not reach spike threshold in any of the cell types not showing burst activity.
Modulation of the afterpotential elicited by three spikes
To further examine the contribution of the ADP to intermittent firing and burst firing, we elicited three spikes at a more depolarized level (Fig. 4). Using constant-current injection, the membrane potential was kept just below the threshold level both in control and in linopirdine. Three 3-ms depolarizing current pulses, with an amplitude sufficient to elicit one spike per pulse, were applied at 50 Hz as shown at the bottom of Fig. 4.
In layer II SCs, each of the three spikes elicited an ADP, which became an AHP after the third spike (Fig. 4, Ai and Aii). The increase of the fast ADP in linopirdine was not enough to trigger a spike on top of the ADP, except for one cell that showed a doublet in control. Traces in control and in linopirdine (30 min) were very similar (Fig. 4Aii), showing almost no effect of the M-current block after three spikes. The total afterpotential area for 30 s after stimulation was almost zero, both in control and in linopirdine (Fig. 4Aiii; n = 8), because the membrane potential was close to the resting level after a short AHP (Fig. 4Aii).
In 25% (3 of 12) of layer II non-SCs, the increase in fast ADP in linopirdine was large enough to reach spike threshold (indicated by an arrow in Fig. 4Bi). These three cells were the cells that showed the three highest intraburst frequencies (80.8 Hz on average) among all layer II non-SCs when their firing pattern was analyzed in the previous section (Fig. 1). This suggests that the increased fast ADP is the mechanism for the high-frequency burst activity of the layer II non-SCs.
The above-mentioned three and the other four layer II non-SCs (7 of 12; 58% of layer II non-SCs) showed another period of firing >500 ms after the stimulation with the application of linopirdine (Fig. 4Bii). We call this firing "delayed firing." Interestingly, the delayed firing resembled the depolarized part of the intermittent firing observed in Fig. 1Bii. Although the duration of delayed firing was usually shorter than that of the depolarization, the firing pattern during delayed firing consisted of doublets and triplets just as observed in Fig. 1B (Fig. 4Bii, inset). Delayed firing was observed even in control conditions in one of the seven cells that showed delayed firing in linopirdine. Three of these seven cells and two other layer II non-SCs (5 of 12; 41% of layer II non-SCs) showed, in control conditions, a slow ADP that peaked >500 ms after stimulation, which corresponds well to the time course of the delayed activity. The afterpotential area, which was nearly zero in control conditions, increased significantly, due to the increase in slow ADP and delayed firing with the application of linopirdine (Fig. 4Biii; n = 12).
In all layer III cells (n = 11), the fast ADP was not sufficient to elicit firing. Layer III cells had a small, slow ADP even under control conditions (Fig. 4Cii, gray trace) and in 17% (2 of 12) of layer III cells the slow ADP was enough to elicit delayed firing. This slow ADP increased and was enough to cause delayed firing in 83% (10 of 12) of cells in linopirdine (Fig. 4Cii, black trace). In contrast to the layer II non-SCs, delayed firing did not consist of high-frequency bursts, but of regular firing. The afterpotential area increased significantly in linopirdine and was by far the largest among all cell types (Fig. 4Ciii; n = 11; note different scaling). This slow and large depolarization, with regular firing on top of it, resembled the depolarized part of the intermittent firing observed in Fig. 1Cii.
In all layer V cells (n = 11), the fast ADP was not enough to cause action potentials and there was no slow ADP (Fig. 4, Di and Dii). They had the largest AHP (in area) among all cell types and the AHP decreased with the application of linopirdine, resulting in a significant increase in the afterpotential area (Fig. 4Diii; n = 11).
In summary, these data indicate that high-frequency bursts in layer II non-SCs can be caused by the increased fast ADP, and the ability of layer II non-SCs and layer III cells to induce slow depolarization might contribute to the formation of the intermittent firing pattern of these cells. In the control studies conducted without linopirdine, no cells showed high-frequency bursts and the slow ADP after three spikes did not develop in any of the cell types; thus no cell type showed a change in afterpotential area (data not shown).
We next examined the spike number dependence and voltage dependence of the delayed firing in layer III cells. Figure 5A illustrates delayed firing elicited by three spikes from –60 mV in linopirdine (30 min). Increasing the spike number to ten produced a larger negative peak between fast and slow ADP, and decreased the amount of delayed firing (Fig. 5B). With 20 spikes, delayed firing did not occur in 83% (5 of 6) of cells tested, causing only an AHP (Fig. 5C). Further increase in the number of elicited spikes caused a larger AHP and no ADP in all six cells tested. In Fig. 5, D and E, three spikes were elicited from hyperpolarized membrane potentials. The slow ADP was smaller at –65 mV (Fig. 5D) and vanished at –70 mV (Fig. 5E), showing that the slow ADP is membrane voltage dependent. In the other four cells that caused delayed firing at above –60 mV, hyperpolarizing the membrane potential to –60 mV was enough to stop delayed firing (data not shown). These results suggest that there is an optimal number (three to five) of spikes and an optimal membrane potential for eliciting delayed firing.
Intrinsic excitability
The role of the M-current in the control of intrinsic excitability of neurons has been an issue of great interest since its discovery. In this study, the effect of the M-current suppression on intrinsic excitability was investigated and compared among the four groups of neurons in the EC (Fig. 6). Neurons were held at –60 mV both in control and in linopirdine (30 min), and a 1-s depolarizing current was injected. The amplitude of the current was chosen to elicit at least eight spikes (ranging from 8 to 15) in the control recordings.
The layer II SCs showed similar firing patterns both in control and in linopirdine under current injections (Fig. 6, Ai and Aii). The number of spikes during the current injections did not increase significantly (Fig. 6Aiii; n = 7).
All layer II non-SCs fired tonically in the control (Fig. 6Bi) and 75% (6 of 8) of them switched to high-frequency burst firing in linopirdine (Fig. 6Bii). The increase in the number of spikes was not significant (Fig. 6Biii; n = 8), which could be due to changes in firing patterns. Both of the two cells that did not switch to high-frequency burst firing showed an increase in the number of spikes (8.5 in control and 11 in linopirdine on average). Figure 6, Biv and Bv shows interspike interval (ISI) histograms of the six layer II non-SCs whose firing pattern switched to high-frequency burst firing in linopirdine. The ISI histogram shows a peak at 100–120 ms in the control condition where the cells were firing tonically (Fig. 6Biv). After application of linopirdine, the ISI histogram shows a sharp peak at 20–40 ms, which corresponds to the intraburst ISI, and a smaller peak at 100–120 ms, which corresponds to the interburst ISI (Fig. 6Bv). Figure 6Bvi shows ISI histograms of all eight layer II non-SCs in the control (open circles) and in linopirdine (filled circles).
The layer III cells showed significantly larger depolarization due to the current injection under linopirdine (Fig. 6, Ci and Cii). The number of spikes increased significantly (Fig. 6Ciii; n = 6), showing the greatest increase in excitability among the four groups. The layer V cells also showed a significant increase in number of spikes (Fig. 6D). These results suggest that an increase in intrinsic excitability depends on the cell type in the EC. In control studies conducted without linopirdine, no cell type showed increased intrinsic excitability. In fact, layer V cells showed a decrease of the number of spikes (data not shown).
Roles of the M-current in subthreshold membrane potential oscillations
Layer II SCs and layer V cells in the EC show subthreshold membrane potential oscillations. Subthreshold membrane potential oscillations in layer II SCs are known to be induced by the interaction between the persistent low-threshold Na+ current (INaP) and the hyperpolarization-activated inward current (Ih; Dickson et al. 2000; Klink and Alonso 1993). In layer V cells, although the involvement of INaP to the subthreshold membrane potential oscillations is clear (Agrawal et al. 2001), the counterpart that is necessary for the generation of subthreshold membrane potential oscillations has not yet been clarified. Because some layer V cells that show no sign of Ih still show robust subthreshold membrane potential oscillations (Hamam et al. 2000), the counterpart has been assumed to be the M-current (Hamam et al. 2002). In this series of experiments, the effects of the M-current blockade on subthreshold membrane potential oscillations were analyzed quantitatively.
Figure 7, Ai and Aii shows subthreshold membrane potential oscillations of the layer II SCs in the control (2.6 ± 0.1 Hz) and 30 min (2.5 ± 0.2 Hz) after linopirdine application, respectively. The membrane potentials were kept constant by injection of a constant current. The subthreshold membrane potential oscillations were clearly seen even 30 min after the application of linopirdine, suggesting that subthreshold membrane potential oscillations in layer II SCs indeed do not depend on the M-current. Power spectra (Fig. 7, Aiii and Aiv) were obtained from short (3.28 s) windows of the membrane potential that corresponded to the section between arrows (Fig. 7, Ai and Aii) in each condition. The peak at 3.4 Hz, which is the main frequency component of the subthreshold membrane potential oscillations for this cell, was kept intact even 30 min after the application of linopirdine. To quantitatively analyze the subthreshold membrane potential oscillations, we obtained three 3.28-s windows of membrane potential traces whose power spectrum had the three largest areas around the peak (1.53–3.97 Hz). Three windows were chosen automatically by a Matlab code in each condition and the areas of the three power spectra were averaged (see METHODS for details). The averaged area around the peak of the power spectrum did not decrease significantly with linopirdine (Fig. 7Av; n = 6).
In contrast, the subthreshold membrane potential oscillations observed in the layer V cells in control (2.4 ± 0.1 Hz) significantly decreased in amplitude after linopirdine application (Fig. 7, Bi and Bii). The sharp peak in the power spectrum was greatly reduced at 30 min after application of linopirdine (Fig. 7, Biii and Biv). The area around the peak of the power spectrum decreased significantly 30 min (n = 7) after linopirdine application (Fig. 7Bv). These results strongly suggest that the subthreshold membrane potential oscillations in layer V cells depend on the M-current. Control studies without linopirdine were conducted in three layer V cells. Subthreshold membrane potential oscillations were not diminished and remained intact in all three cells, even 45 min after onset of the recordings. Areas of the power spectrum were: control, 0.32 ± 0.16 mV2; 45 min, 0.31 ± 0.15 mV2.
Effect of XE991 and SK-channel blocker
It is reported that linopirdine moderately inhibits the IAHP, which could result from small-conductance calcium-activated potassium (SK) current (Schnee and Brown 1998). We thus tested XE991, an M-current blocker that is more specific to M-channels, along with the SK-channel blocker apamin, to confirm that the results obtained with linopirdine were not caused simply by inhibition of the SK current.
Using layer V cells, we first tested the single-spike protocol used in Fig. 2. Bath application of apamin (300 nM) for 15 min did not cause a significant change in any of the measurements of ADP and AHP in all layer V cells (Fig. 8A, compare black and blue traces; n = 9). We then bath-applied XE991 (10 µM) and apamin (300 nM) at the same time and waited for 30 min. This increased all measurements of ADP (amplitude, duration, and area) and significantly decreased AHP amplitudes (Fig. 8A, compare blue and red traces; n = 9). We encountered two cells where the ADP was large enough to reach the firing threshold and caused a high-frequency burst when XE991 and apamin were applied at the same time (Fig. 8B). These results indicate these two cells had a greater effect of SK current and M-current block on the size of ADP. However, because the response was qualitatively different from that of the other nine cells, these two cells were not included in the statistical analysis (Fig. 8, Aiii–Avi, Ciii, and Div).
We next tested the three-spike protocol used in Fig. 4. Although application of apamin did not significantly increase the afterpotential area, further application of XE991 along with apamin significantly increased the afterpotential area (Fig. 8C; n = 9).
Intrinsic excitability was measured in the same way as in Fig. 6. Although application of apamin did not significantly increase the number of spikes, further application of XE991 along with apamin significantly increased the number of spikes (Fig. 8D; n = 9).
To further show that intermittent firing and delayed firing of layer III cells are not dependent on weak selectivity of linopirdine, we tested XE991 in four layer III cells. We applied 5 µM XE991 in two cells and 10 µM XE991 in the other two cells for 30 min. Both of the two cells in 5 µM and 10 µM XE991 caused intermittent firing (Fig. 9, A and B), developed slow ADP, and showed delayed firing (Fig. 9C). These experiments strongly suggest that the increase of ADP, decrease of AHP, development of slow ADP, and intermittent firing patterns that were observed with linopirdine were not due to inhibition of ionic currents other than the M-current.
|
|
|
DISCUSSION |
|---|
|
; Shah et al. 2002), others consider that the Kv7.2 homomers are not included in the M-current (Schwarz et al. 2006). Slowly activated and deactivating K+ current (IKs) in nodes of Ranvier is encoded by Kv7.2 homomers and is suppressed by both linopirdine and XE991 with lower affinities compared with those for Kv7.2/Kv7.3 heteromers. It is therefore possible that the block of Kv7.2 homomers (nodal IKs) is partly contributing to the result obtained herein. We have shown that blockade of the M-current causes intermittent firing in layer II non-SCs and layer III cells. Whereas layer II non-SCs showed high-frequency burst firing during intermittent firing, layer III cells showed regular firing during intermittent firing (Fig. 1). The single-spike afterpotential at –60 mV did not cause burst firing in any cell type, whereas layer III cells showed large, slow ADPs (Fig. 2). When spikes were elicited at depolarized potentials, the increased fast ADP in layer II non-SCs was large enough to cause burst firing, suggesting a contribution of fast ADP to high-frequency bursting (Fig. 4). The afterpotential after three spikes also revealed delayed firing in layer II non-SCs and layer III cells, suggesting that the modulation of the M-current played a role in their delayed firing activity. The increase in excitability was most pronounced in the layer III cells (Fig. 6). The subthreshold membrane potential oscillations in layer V cells were greatly suppressed, whereas those in the layer II SCs remained intact after blockade of the M-current (Fig. 7). The M-current thus shaped the firing behavior, intrinsic excitability, and subthreshold membrane potential oscillations of principal cells in the EC in a cell-type–specific manner. In this section, we discuss how these modifications would contribute to the physiological functions of the EC.
Induction of burst firing by M-current suppression
In the present work, the layer II non-SCs showed high-frequency (98 Hz) burst firing (Fig. 1Bii). Although ADP did not reach the spike threshold when we tested cells at a membrane potential of –60 mV (Fig. 2B), firing at the peak of the ADP was observed at the depolarized level in three cells (Fig. 4Bi). This indicates that ADP in EC layer II non-SCs is indeed large enough to reach firing threshold at depolarized levels.
Alonso and García-Austt (1987) observed firing patterns of EC neurons during theta rhythm activity. Interestingly, they showed two classes of rhythmic cells in the layer II. The class I rhythmic cells showed strong modulation to the theta rhythm and a small tendency to burst. The class II rhythmic cells showed weak modulation to the theta rhythm and a strong tendency to burst. These two classes of cells are implicative of SCs and non-SCs, respectively. In our data, the SCs showed intact subthreshold membrane potential oscillations and no burst firing, and the non-SCs showed no clear subthreshold membrane potential oscillations, and burst firing, under the blockade of the M-current. These different tendencies with regard to burst firing also agree with a previous observation by Klink and Alonso (1997). They showed, in an EC slice preparation, that only non-SCs show burst firing with a bath application of carbachol. As mentioned earlier, the layer II non-SCs project mainly to the dentate gyrus (DG) and hippocampal CA3 regions through the perforant pathway (Schwartz and Coleman 1981; Tamamaki and Nojyo 1993; Witter and Amaral 1991). Because spike bursts facilitate synaptic plasticity (Magee and Johnston 1997; Thomas et al. 1998), high-frequency bursts of layer II non-SCs may help not only to drive neurons in the DG and CA3, but also to store memories during cholinergic activation.
ADP increase and AHP decrease during M-current suppression
The spike ADP results from currents through T-, R-, and L-type voltage-gated Ca2+ channels, which are mainly distributed in distal dendrites (Golding et al. 1999; Magee and Carruth 1999; Metz et al. 2005; White et al. 1989; Wong and Prince 1981), and persistent Na+ channels, which are mainly distributed proximal to soma (Azouz et al. 1996; Yue et al. 2005). Although it is not clear which of the two types of currents contributes most to the ADP observed in neurons in the EC, both of them have been shown to be under the control of K+ currents. Suppression of A-type K+ currents at distal dendrites increases Ca2+-mediated ADP (Magee and Carruth 1999) and suppression of the M-current at the soma increases Na+-mediated ADP (Yue and Yaari 2006), and both of these cause burst firing. It has also been shown that under A-type K+-current suppression, the M-current contributes to the enhancement of Ca2+ spike generation in dendrites, suggesting that the M-current can modulate both somatic and dendritic sources of the ADP (Yue and Yaari 2006).
In the present study, the application of linopirdine increased the ADP and decreased the AHP except in the case of layer II SCs (Fig. 2). We further suggested that the increase in ADP is not due to the lack of M-current that is activated by an action potential (Fig. 3). The time constant for activation of the M-current (6 to 50 ms depending on the membrane potential; Brown and Adams 1980; Wang et al. 1998) is too long for an action potential to activate the M-current. However, when an action potential is elicited at –60 mV, although the action potential itself lasts for only a few milliseconds, the ADP lasts for tens of milliseconds, keeping the membrane potential well above the threshold for the M-current activation. Thus the ADP itself, whether it has dendritic or somatic origin, would activate the M-current in normal conditions. Once activated by the ADP, the M-current takes time to deactivate and, during this time, it probably contributes to the AHP. Block of the M-current would thus "unleash" development of the ADP (Yue and Yaari 2006) and decrease the contribution to the AHP.
Ionic currents underlying delayed firing
In the present study, layer II non-SCs and layer III cells showed delayed firing (Fig. 4, B and C). Delayed firing was observed previously in layer II SCs and non-SCs during cholinergic activation with carbachol (Klink and Alonso 1997; Magistretti et al. 2004). It is believed to play an important role in delayed matching tasks and may be the underlying mechanism for sustained spiking activity (Fransén et al. 2002). Delayed firing is caused by depolarization by a nonspecific cation current (INCM) that is activated by the muscarinic receptor (Shalinsky et al. 2002). The INCM has transient tail and sustained plateau components (Magistretti et al. 2004). The tail part of this current is responsible for the transient depolarization during delayed firing and is sensitive to Ca2+ influx from the voltage-gated Ca2+ channels. The time course of the tail current fits well with that of delayed firings observed in this study. The INCM in layer II cells has many analogies with currents mediated by transient receptor potential (TRP) channels (Shalinsky et al. 2002), and a recent study in EC layer V cells suggests that the INCM is mediated by TRP channels (Tahvildari et al. 2004). Many TRP channels display constitutive activity (Pedersen et al. 2005), suggesting that some INCM channels are active without muscarinic activation. Based on this, we propose that the delayed firing in our experiments was caused by the tail component of constitutive INCM. Three spikes elicited at 50 Hz caused larger Ca2+ influx than a single spike and induced INCM tail. Indeed, some of the layer II non-SCs and layer III cells had a small slow ADP even before applying linopirdine. Application of linopirdine, which eliminated the "clamping effect," enabled the relatively small and constitutively active INCM tail current to drive delayed firing.
Carbachol induced persistent firing in layer V cells in the medial EC (Egorov et al. 2002) and layer III cells in the lateral EC (Tahvildari and Alonso 2005) due to INCM. Persistent firing is believed to be crucial for mnemonic activities of the EC (McGaughy et al. 2005). In the present study, extended periods of persistent firing were never observed, even though different numbers of elicited spikes were tested (Fig. 5, A–C). Rather, the increased number of spikes caused a larger AHP and less delayed firing. This shows that the INCM current, which is constitutively active, is significantly smaller than the one during muscarinic activation, resulting in domination by the AHP current. In Fig. 5, D and E, we have shown voltage dependence of the delayed firing. Delayed firing and the slow ADP were not observed at hyperpolarized membrane potentials. Very similar voltage dependence was observed in layer V cells in carbachol (Egorov et al. 2002), where stimulation causing persistent firing at –60 mV caused only delayed firing at around –70 mV and no depolarization at all at around –80 mV. This observation thus supports the idea that delayed firing is caused by INCM. Nevertheless, suppression of the M-current during cholinergic activation would help INCM to depolarize the membrane potential to trigger delayed firing in layer II cells, or to initiate and maintain (see following text) persistent activity in layer III cells.
Modulation of excitability
Effects of the M-current on intrinsic excitability have been investigated since early studies by Adams et al. (1982a,b). Increased excitability was often measured as reduced spike frequency adaptation (Aiken et al. 1995; Cole and Nicoll 1983; Madison and Nicoll 1984; McCormick and Williamson 1989; Peters et al. 2005; Wang et al. 1998), whereas in some preparations, reduction of spike frequency adaptation was not observed as a result of M-current blockade (Miles et al. 2005; Romero et al. 2004). In the EC, whereas layer III and layer V cells showed a clear reduction of spike frequency adaptation, layer II SCs and non-SCs did not show a significant change (Fig. 6). Layer II SCs in particular showed the least dependence on the M-current, with no significant change in their firing patterns (Fig. 1A), ADP area and AHP amplitude (Fig. 2A), delayed firing properties (Fig. 4A), or ability to produce subthreshold membrane potential oscillations (Fig. 7). One of the characteristics of the layer II SCs is their huge Ih conductance. When the membrane potential is hyperpolarized, Ih is activated and the resulting inward current depolarizes the membrane. On the other hand, when the membrane is depolarized, Ih is deactivated and the reduced inward current hyperpolarizes the membrane; Ih thus also contributes to the "clamping effect." Layer II SCs, having the largest Ih conductance among all cell types compared here, would still have a substantial "clamping effect," even under blockade of the M-current, resulting in the smallest contribution of the M-current on excitability. Layer III cells, which have the least Ih conductance (in the soma), showed the largest increase in excitability (Fig. 6C).
Mutations of M-channel (Kv7.2 and Kv7.3) genes lead to benign familial neonatal convulsions (BFNC), a predominantly inherited epilepsy (Biervert et al. 1998; Charlier et al. 1998; Singh et al. 1998). The finding that hyperexcitability in BFNC seems to be caused by only a 25% reduction of the M-current (Schroeder et al. 1998) revealed the significant contribution of the M-current in the maintenance of proper excitability. The EC is often the source of temporal lobe epilepsy (TLE; Bartolomei et al. 2005) and hyperexcitability during epilepsy causes neuronal loss. Within the EC, neuronal loss is preponderant in layer III after TLE, whereas layer II cells are relatively resistant (Du et al. 1993, 1995; Scharfman 2000; Schwarcz and Witter 2002). Interestingly, our results showed that the increase in excitability was greatest in layer III cells and smallest in layer II cells after linopirdine treatment (Fig. 6). Furthermore, whereas the membrane potential of layer III and layer V cells depolarized about 12 mV, those in layer II cells depolarized <6 mV in linopirdine. This consistency of a pattern, linking cell loss in TLE and sensitivity of the intrinsic excitability to the M-current block, suggests that hyperexcitability, resulting from the suppression of M-current during TLE, could contribute to the cell loss in layer III.
Increased excitability would also contribute to persistent firing of EC cells (Egorov et al. 2002; Tahvildari and Alonso 2005) under muscarinic activation. During persistent firing, membrane potentials of neurons are kept depolarized at a suprathreshold level. In this study a significant increase in excitability was observed at this level under the M-current block. Increased excitability would greatly assist INCM in keeping the membrane potential above the threshold. The M-current would thus help both the initiation and the maintenance of persistent firing of neurons in the EC. Indeed, layer III and layer V cells, which showed a significant increase in excitability, are the neurons that show persistent firing.
Modulation of subthreshold membrane potential oscillations
Underlying ionic currents for subthreshold membrane potential oscillations vary in different neuronal populations (Alonso and Llinás 1989; Gutfreund et al. 1995; Pape et al. 1998; Wang 1993). The subthreshold membrane potential oscillations in layer II SCs has been shown to be generated by INaP and Ih (Dickson et al. 2000; Klink and Alonso 1993). As for layer V cells, Agrawal et al. (2001) showed that subthreshold membrane potential oscillations depend on INaP. However, the other current that interacts with INaP remained unknown. The observation that there are layer V cells that have no sign of Ih (hyperpolarization induced sag) and still show robust subthreshold membrane potential oscillations (Hamam et al. 2000) suggested that the mechanism for subthreshold membrane potential oscillations of layer V cells is different from that for layer II SCs. Gutfreund et al. (1995) showed that subthreshold membrane potential oscillations in guinea pig frontal cortex neurons can be replicated by INaP and a noninactivating potassium current with the kinetics of the M-current, suggesting that this is also the case in the EC layer V cells (Hamam et al. 2002). Our results clearly show that blockade of the M-current did not affect subthreshold membrane potential oscillations in layer II SCs, but suppressed the subthreshold membrane potential oscillations in layer V cells (Fig. 7), showing that subthreshold membrane potential oscillations in layer V cells indeed depend on the M-current.
That cholinergic modulation differently affects the subthreshold membrane potential oscillations in layer II and layer V cells is interesting because cholinergic modulation is believed to set the functional state of the entorhinal–hippocampal network. A high acetylcholine level promotes encoding of information in the hippocampus, whereas a low acetylcholine level promotes consolidation of information from the hippocampus into neocortical regions (Hasselmo 1999). In the encoding phase, the EC layer II serves as the main source of information to the hippocampus through the perforant pathway. In this phase, theta rhythm, crucial for temporal coding and synaptic modification (Buzsáki 2002), is observed in the EC and the hippocampus. In the EC, neurons in the superficial layers (layers II and III) fire in synchrony with the theta rhythm, whereas most of the neurons from deep layers (layers V and VI) do not (Alonso and García-Austt 1987; Chrobak and Buzsáki 1994; Frank et al. 2001). For generation of the theta rhythm, input from septal fibers plays a crucial role. Intrinsic oscillatory properties of the EC cells, however, may also contribute to the generation of the theta rhythm, as is believed to be the case in the hippocampus (Buzsáki 2002). If this is true, the subthreshold membrane potential oscillations that remain intact in layer II SCs, and vanish in layer V cells under suppression of the M-current, will contribute to the layer-dependent properties of the theta rhythm activity during mnemonic processes.
In conclusion, modulation of the M-current has the potential to help shape the dynamical properties of neurons in the EC, such as burst firing patterns, delayed firing, persistent firing, and oscillation properties, all of which are important components of memory function. This tuning, performed in a layer- and cell-type–specific manner, is in keeping with the transition of the entorhinal–hippocampal area to the encoding stage. This study suggests that the M-current is therefore not simply a controller of excitability, but is a key component that tunes the dynamics of neurons for mnemonic function.
|
|
GRANTS |
|---|
|
|
|
ACKNOWLEDGMENTS |
|---|
|
Present address of M. Yoshida: Center for Memory and Brain, Boston University, 2 Cummington Street, Boston, MA 02215.
|
|
FOOTNOTES |
|---|
Deceased July 6, 2005. 
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Address for reprint requests and other correspondence: M. Yoshida, Center for Memory and Brain, Boston University, 2 Cummington Street, Boston, MA 02215 (E-mail: motoharu{at}bu.edu)
|
|
REFERENCES |
|---|
|
Adams PR, Brown DA, Constanti A. Pharmacological inhibition of the M-current. J Physiol 332: 223–262, 1982b.
Agrawal N, Hamam BN, Magistretti J, Alonso A, Ragsdale DS. Persistent sodium channel activity mediates subthreshold membrane potential oscillations and low-threshold spikes in rat entorhinal cortex layer V neurons. Neuroscience 102: 53–64, 2001.[CrossRef][Web of Science][Medline]
Aiken SP, Lampe BJ, Murphy PA, Brown BS. Reduction of spike frequency adaptation and blockade of M-current in rat CA1 pyramidal neurones by linopirdine (DuP 996), a neurotransmitter release enhancer. Br J Pharmacol 115: 1163–1168, 1995.[Web of Science][Medline]
Alonso A, Garcia-Austt E. Neuronal sources of theta rhythm in the entorhinal cortex of the rat. II. Phase relations between unit discharges and theta field potentials. Exp Brain Res 67: 502–509, 1987.[Web of Science][Medline]
Alonso A, Klink R. Differential electroresponsiveness of stellate and pyramidal-like cells of medial entorhinal cortex layer II. J Neurophysiol 70: 128–143, 1993.
Alonso A, Llinás RR. Subthreshold Na+-dependent theta-like rhythmicity in stellate cells of entorhinal cortex layer II. Nature 342: 175–177, 1989.[CrossRef][Medline]
Alonso JR, U HS, Amaral DG. Cholinergic innervation of the primate hippocampal formation: II. Effects of fimbria/fornix transection. J Comp Neurol 375: 527–551, 1996.[CrossRef][Web of Science][Medline]
Azouz R, Jensen MS, Yaari Y. Ionic basis of spike after-depolarization and burst generation in adult rat hippocampal CA1 pyramidal cells. J Physiol 492: 211–223, 1996.
Bartolomei F, Khalil M, Wendling F, Sontheimer A, Regis J, Ranjeva JP, Guye M, Chauvel P. Entorhinal cortex involvement in human mesial temporal lobe epilepsy: an electrophysiologic and volumetric study. Epilepsia 46: 677–687, 2005.[CrossRef][Web of Science][Medline]
Biervert C, Schroeder BC, Kubisch C, Berkovic SF, Propping P, Jentsch TJ, Steinlein OK. A potassium channel mutation in neonatal human epilepsy. Science 279: 403–406, 1998.
Brown D. M-currents: an update. Trends Neurosci 11: 294–299, 1988.[CrossRef][Web of Science][Medline]
Brown DA, Adams PR. Muscarinic suppression of a novel voltage-sensitive K+ current in a vertebrate neurone. Nature 283: 673–676, 1980.[CrossRef][Medline]
Brown DA, Hughes SA, Marsh SJ, Tinker A. Regulation of M(Kv7.2/7.3) channels in neurons by PIP2 and products of PIP2 hydrolysis: significance for receptor-mediated inhibition. J Physiol 582: 917–925, 2007.
Brown DA, Selyanko AA, Hadley JK, Tatulian L. Some pharmacological properties of neural KCNQ channels. Neurophysiology 34: 91–94, 2002.[CrossRef]
Buckmaster CA, Eichenbaum H, Amaral DG, Suzuki WA. Entorhinal cortex lesions disrupt the relational organization of memory in monkeys. J Neurosci 24: 9811–9825, 2004.
Burwell RD, Amaral DG. Cortical afferents of the perirhinal, postrhinal, and entorhinal cortices of the rat. J Comp Neurol 398: 179–205, 1998a.[CrossRef][Web of Science][Medline]
Burwell RD, Amaral DG. Perirhinal and postrhinal cortices of the rat: interconnectivity and connections with the entorhinal cortex. J Comp Neurol 391: 293–321, 1998b.[CrossRef][Web of Science][Medline]
Buzsáki G. Theta oscillations in the hippocampus. Neuron 33: 325–340, 2002.[CrossRef][Web of Science][Medline]
Charlier C, Singh NA, Ryan SG, Lewis TB, Reus BE, Leach RJ, Leppert M. A pore mutation in a novel KQT-like potassium channel gene in an idiopathic epilepsy family. Nat Genet 18: 53–55, 1998.[CrossRef][Web of Science][Medline]
Chrobak JJ, Buzsáki G. Selective activation of deep layer (v–vi) retrohippocampal cortical neurons during hippocampal sharp waves in the behaving rat. J Neurosci 14: 6160–6170, 1994.[Abstract]
Cole AE, Nicoll RA. Acetylcholine mediates a slow synaptic potential in hippocampal pyramidal cells. Science 221: 1299–1301, 1983.
Cooper EC, Harrington E, Jan YN, Jan LY. M channel KCNQ2 subunits are localized to key sites for control of neuronal network oscillations and synchronization in mouse brain. J Neurosci 21: 9529–9540, 2001.
Costa AM, Brown BS. Inhibition of M-current in cultured rat superior cervical ganglia by linopirdine: mechanism of action studies. Neuropharmacology 36: 1747–1753, 1997.[CrossRef][Web of Science][Medline]
Dickson CT, Magistretti J, Shalinsky MH, Fransen E, Hasselmo ME, Alonso A. Properties and role of I(h) in the pacing of subthreshold oscillations in entorhinal cortex layer II neurons. J Neurophysiol 83: 2562–2579, 2000.
Dickson CT, Mena AR, Alonso A. Electroresponsiveness of medial entorhinal cortex layer III neurons in vitro. Neuroscience 81: 937–950, 1997.[CrossRef][Web of Science][Medline]
Du F, Eid T, Lothman EW, Kohler C, Schwarcz R. Preferential neuronal loss in layer III of the medial entorhinal cortex in rat models of temporal lobe epilepsy. J Neurosci 15: 6301–6313, 1995.
Du F, Whetsell WO Jr, Abou-Khalil B, Blumenkopf B, Lothman EW, Schwarcz R. Preferential neuronal loss in layer III of the entorhinal cortex in patients with temporal lobe epilepsy. Epilepsy Res 16: 223–233, 1993.[CrossRef][Web of Science][Medline]
Egorov AV, Hamam BN, Fransen E, Hasselmo ME, Alonso AA. Graded persistent activity in entorhinal cortex neurons. Nature 420: 173–178, 2002.[CrossRef][Medline]
Frank LM, Brown EN, Wilson MA. A comparison of the firing properties of putative excitatory and inhibitory neurons from CA1 and the entorhinal cortex. J Neurophysiol 86: 2029–2040, 2001.
Fransén E, Alonso AA, Dickson CT, Magistretti J, Hasselmo ME. Ionic mechanisms in the generation of subthreshold oscillations and action potential clustering in entorhinal layer II stellate neurons. Hippocampus 14: 368–384, 2004.[CrossRef][Web of Science][Medline]
Fransén E, Alonso AA, Hasselmo ME. Simulations of the role of the muscarinic-activated calcium-sensitive nonspecific cation current INCM in entorhinal neuronal activity during delayed matching tasks. J Neurosci 22: 1081–1097, 2002.
Fransén E, Wallenstein GV, Alonso A, Dickson CT, Hasselmo ME. A biophysical simulation of intrinsic and network properties of entorhinal cortex. Neurocomputing 26–27: 375–380, 1999.[CrossRef]
Golding NL, Jung HY, Mickus T, Spruston N. Dendritic calcium spike initiation and repolarization are controlled by distinct potassium channel subtypes in CA1 pyramidal neurons. J Neurosci 19: 8789–8798, 1999.
Gu N, Vervaeke K, Hu H, Storm JF. Kv7/KCNQ/M and HCN/h, but not KCa2/SK channels, contribute to the somatic medium after-hyperpolarization and excitability control in CA1 hippocampal pyramidal cells. J Physiol 566: 689–715, 2005.
Gutfreund Y, Yarom Y, Segev I. Subthreshold oscillations and resonant frequency in guinea-pig cortical neurons: physiology and modelling. J Physiol 483: 621–640, 1995.
Hadley JK, Passmore GM, Tatulian L, Al-Qatari M, Ye F, Wickenden AD, Brown DA. Stoichiometry of expressed KCNQ2/KCNQ3 potassium channels and subunit composition of native ganglionic M channels deduced from block by tetraethylammonium. J Neurosci 23: 5012–5019, 2003.
Hamam BN, Amaral DG, Alonso AA. Morphological and electrophysiological characteristics of layer V neurons of the rat lateral entorhinal cortex. J Comp Neurol 451: 45–61, 2002.[CrossRef][Web of Science][Medline]
Hamam BN, Kennedy TE, Alonso A, Amaral DG. Morphological and electrophysiological characteristics of layer V neurons of the rat medial entorhinal cortex. J Comp Neurol 418: 457–472, 2000.[CrossRef][Web of Science][Medline]
Hasselmo ME. Neuromodulation: acetylcholine and memory consolidation. Trends Cogn Sci 3: 351–359, 1999.[CrossRef][Web of Science][Medline]
Hu H, Vervaeke K, Storm JF. Two forms of electrical resonance at theta frequencies, generated by M-current, h-current and persistent Na+ current in rat hippocampal pyramidal cells. J Physiol 545: 783–805, 2002.
Insausti R, Amaral DG, Cowan WM. The entorhinal cortex of the monkey: II. Cortical afferents. J Comp Neurol 264: 356–395, 1987.[CrossRef][Web of Science][Medline]
Insausti R, Herrero MT, Witter MP. Entorhinal cortex of the rat: cytoarchitectonic subdivisions and the origin and distribution of cortical efferents. Hippocampus 7: 146–183, 1997.[CrossRef][Web of Science][Medline]
Klink R, Alonso A. Ionic mechanisms for the subthreshold oscillations and differential electroresponsiveness of medial entorhinal cortex layer II neurons. J Neurophysiol 70: 144–157, 1993.
Klink R, Alonso A. Muscarinic modulation of the oscillatory and repetitive firing properties of entorhinal cortex layer II neurons. J Neurophysiol 77: 1813–1828, 1997.
Koyama S, Appel SB. Characterization of M-current in ventral tegmental area dopamine neurons. J Neurophysiol 96: 535–543, 2006.
Lamas JA, Selyanko AA, Brown DA. Effects of a cognition-enhancer, linopirdine (DuP 996), on M-type potassium currents (IK(M)) and some other voltage- and ligand-gated membrane currents in rat sympathetic neurons. Eur J Neurosci 9: 605–616, 1997.[CrossRef][Web of Science][Medline]
Leonard BW, Amaral DG, Squire LR, Zola-Morgan S. Transient memory impairment in monkeys with bilateral lesions of the entorhinal cortex. J Neurosci 15: 5637–5659, 1995.[Abstract]
Madison DV, Nicoll RA. Control of the repetitive discharge of rat CA 1 pyramidal neurones in vitro. J Physiol 354: 319–331, 1984.
Magee JC, Carruth M. Dendritic voltage-gated ion channels regulate the action potential firing mode of hippocampal CA1 pyramidal neurons. J Neurophysiol 82: 1895–1901, 1999.
Magee JC, Johnston D. A synaptically controlled, associative signal for Hebbian plasticity in hippocampal neurons. Science 275: 209–213, 1997.
Magistretti J, Ma L, Shalinsky MH, Lin W, Klink R, Alonso A. Spike patterning by Ca2+-dependent regulation of a muscarinic cation current in entorhinal cortex layer II neurons. J Neurophysiol 92: 1644–1657, 2004.
McCormick DA, Williamson A. Convergence and divergence of neurotransmitter action in human cerebral cortex. Proc Natl Acad Sci USA 86: 8098–8102, 1989.
McGaughy J, Koene RA, Eichenbaum H, Hasselmo ME. Cholinergic deafferentation of the entorhinal cortex in rats impairs encoding of novel but not familiar stimuli in a delayed nonmatch-to-sample task. J Neurosci 25: 10273–10281, 2005.
Metz AE, Jarsky T, Martina M, Spruston N. R-type calcium channels contribute to afterdepolarization and bursting in hippocampal CA1 pyramidal neurons. J Neurosci 25: 5763–5773, 2005.
Miles GB, Dai Y, Brownstone RM. Mechanisms underlying the early phase of spike frequency adaptation in mouse spinal motoneurones. J Physiol 566: 519–532, 2005.
Naber PA, Lopes da Silva FH, Witter MP. Reciprocal connections between the entorhinal cortex and hippocampal fields CA1 and the subiculum are in register with the projections from CA1 to the subiculum. Hippocampus 11: 99–104, 2001.[CrossRef][Web of Science][Medline]
Pape HC, Paré D, Driesang RB. Two types of intrinsic oscillations in neurons of the lateral and basolateral nuclei of the amygdala. J Neurophysiol 79: 205–216, 1998.
Pedersen SF, Owsianik G, Nilius B. TRP channels: an overview. Cell Calcium 38: 233–252, 2005.[CrossRef][Web of Science][Medline]
Peters HC, Hu H, Pongs O, Storm JF, Isbrandt D. Conditional transgenic suppression of M channels in mouse brain reveals functions in neuronal excitability, resonance and behavior. Nat Neurosci 8: 51–60, 2005.[CrossRef][Web of Science][Medline]
Romero M, Reboreda A, Sanchez E, Lamas JA. Newly developed blockers of the M-current do not reduce spike frequency adaptation in cultured mouse sympathetic neurons. Eur J Neurosci 19: 2693–2702, 2004.[CrossRef][Web of Science][Medline]
Scharfman HE. Epileptogenesis in the parahippocampal region. Parallels with the dentate gyrus. Ann NY Acad Sci 911: 305–327, 2000.[Web of Science][Medline]
Schnee ME, Brown BS. Selectivity of linopirdine (DuP 996), a neurotransmitter release enhancer, in blocking voltage-dependent and calcium-activated potassium currents in hippocampal neurons. J Pharmacol Exp Ther 286: 709–717, 1998.
Schroeder BC, Kubisch C, Stein V, Jentsch TJ. Moderate loss of function of cyclic-AMP-modulated KCNQ2/KCNQ3 K+ channels causes epilepsy. Nature 396: 687–690, 1998.[CrossRef][Medline]
Schwarcz R, Witter MP. Memory impairment in temporal lobe epilepsy: the role of entorhinal lesions. Epilepsy Res 50: 161–177, 2002.[CrossRef][Web of Science][Medline]
Schwartz SP, Coleman PD. Neurons of origin of the perforant path. Exp Neurol 74: 305–312, 1981.[CrossRef][Web of Science][Medline]
Scoville WB, Milner B. Loss of recent memory after bilateral hippocampal lesions. J Neurol Neurosurg Psychiatry 20: 11–21, 1957.
Shah MM, Mistry M, Marsh SJ, Brown DA, Delmas P. Molecular correlates of the M-current in cultured rat hippocampal neurons. J Physiol 544: 29–37, 2002.
Shalinsky MH, Magistretti J, Ma L, Alonso AA. Muscarinic activation of a cation current and associated current noise in entorhinal-cortex layer-II neurons. J Neurophysiol 88: 1197–1211, 2002.
Singh NA, Charlier C, Stauffer D, DuPont BR, Leach RJ, Melis R, Ronen GM, Bjerre I, Quattlebaum T, Murphy JV, McHarg ML, Gagnon D, Rosales TO, Peiffer A, Anderson VE, Leppert M. A novel potassium channel gene, KCNQ2, is mutated in an inherited epilepsy of newborns. Nat Genet 18: 25–29, 1998.[CrossRef][Web of Science][Medline]
Squire LR, Zola-Morgan S. The medial temporal lobe memory system. Science 253: 1380–1386, 1991.
Steffenach HA, Witter M, Moser MB, Moser EI. Spatial memory in the rat requires the dorsolateral band of the entorhinal cortex. Neuron 45: 301–313, 2005.[CrossRef][Web of Science][Medline]
Steriade M, Nunez A, Amzica F. A novel slow (<1 Hz) oscillation of neocortical neurons in vivo: depolarizing and hyperpolarizing components. J Neurosci 13: 3252–3265, 1993.[Abstract]
Storm JF. An after-hyperpolarization of medium duration in rat hippocampal pyramidal cells. J Physiol 409: 171–190, 1989.
Suzuki WA, Miller EK, Desimone R. Object and place memory in the macaque entorhinal cortex. J Neurophysiol 78: 1062–1081, 1997.
Tahvildari B, Alonso A. Switching-on and -off persistent activity in entorhinal cortex layer III pyramidal cells; an intrinsic neuronal toggle mechanism. Program No. 737.7. 2005 Abstract Viewer and Itinerary Planner. Washington, DC: Society for Neuroscience, 2005. Online.
Tahvildari B, Egorov A, Klink R, Alonso A. Persistent activity in the entorhinal cortex neurons, TRP channels and intracellular calcium stores. Program No. 516.3. 2004 Abstract Viewer and Itinerary Planner. Washington, DC: Society for Neuroscience, 2004. Online.
Tamamaki N, Nojyo Y. Projection of the entorhinal layer II neurons in the rat as revealed by intracellular pressure-injection of neurobiotin. Hippocampus 3: 471–480, 1993.[CrossRef][Web of Science][Medline]
Thomas MJ, Watabe AM, Moody TD, Makhinson M, O'Dell TJ. Postsynaptic complex spike bursting enables the induction of LTP by theta frequency synaptic stimulation. J Neurosci 18: 7118–7126, 1998.
Van Groen T, Lopes da Silva FH. Organization of the reciprocal connections between the subiculum and the entorhinal cortex in the cat: II. An electrophysiological study. J Comp Neurol 251: 111–120, 1986.[CrossRef][Web of Science][Medline]
Van Groen T, van Haren FJ, Witter MP, Groenewegen HJ. The organization of the reciprocal connections between the subiculum and the entorhinal cortex in the cat: I. A neuroanatomical tracing study. J Comp Neurol 250: 485–497, 1986.[CrossRef][Web of Science][Medline]
Wang HS, Pan Z, Shi W, Brown BS, Wymore RS, Cohen IS, Dixon JE, McKinnon D. KCNQ2 and KCNQ3 potassium channel subunits: molecular correlates of the M-channel. Science 282: 1890–1893, 1988.[CrossRef]
Wang XJ. Ionic basis for intrinsic 40 Hz neuronal oscillations. Neuroreport 5: 221–224, 1993.[Web of Science][Medline]
White G, Lovinger DM, Weight FF. Transient low-threshold Ca2+ current triggers burst firing through an afterdepolarizing potential in an adult mammalian neuron. Proc Natl Acad Sci USA 86: 6802–6806, 1989.
Witter MP, Amaral DG. Entorhinal cortex of the monkey: V. Projections to the dentate gyrus, hippocampus, and subicular complex. J Comp Neurol 307: 437–459, 1991.[CrossRef][Web of Science][Medline]
Wong RK, Prince DA. Afterpotential generation in hippocampal pyramidal cells. J Neurophysiol 45: 86–97, 1981.
Yoshida M, Alonso A. Differential role of the M-current on the oscillatory and bursting behavior of principal cells from layers II, III and V of the entorhinal cortex. Program No. 971.5. 2005 Abstract Viewer and Itinerary Planner. Washington, DC: Society for Neuroscience, 2005. Online.
Yue C, Remy S, Su H, Beck H, Yaari Y. Proximal persistent Na+ channels drive spike afterdepolarizations and associated bursting in adult CA1 pyramidal cells. J Neurosci 25: 9704–9720, 2005.
Yue C, Yaari Y. KCNQ/M channels control spike afterdepolarization and burst generation in hippocampal neurons. J Neurosci 24: 4614–4624, 2004.
Yue C, Yaari Y. Axo-somatic and apical dendritic Kv7/M channels differentially regulate the intrinsic excitability of adult rat CA1 pyramidal cells. J Neurophysiol 95: 3480–3495, 2006.
Zaczek R, Chorvat RJ, Saye JA, Pierdomenico ME, Maciag CM, Logue AR, Fisher BN, Rominger DH, Earl RA. Two new potent neurotransmitter release enhancers, 10,10-bis(4-pyridinylmethyl)-9(10H)-anthracenone and 10,10-bis(2-fluoro-4-pyridinylmethyl)-9(10H)-anthracenone: comparison to linopirdine. J Pharmacol Exp Ther 285: 724–730, 1998.
This article has been cited by other articles:
|
|
 |
|
  M. G. Sheroziya, O. von Bohlen und Halbach, K. Unsicker, and A. V. Egorov Spontaneous Bursting Activity in the Developing Entorhinal Cortex J. Neurosci., September 30, 2009; 29(39): 12131 - 12144. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. T. Paz, Sév. Mahon, P. Tiret, S.ép. Genet, B. Delord, and S.ép. Charpier Multiple forms of activity-dependent intrinsic plasticity in layer V cortical neurones in vivo J. Physiol., July 1, 2009; 587(13): 3189 - 3205. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. R. Fernandez and J. A. White Reduction of Spike Afterdepolarization by Increased Leak Conductance Alters Interspike Interval Variability J. Neurosci., January 28, 2009; 29(4): 973 - 986. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. G. Burton, M. N. Economo, G. J. Lee, and J. A. White Development of Theta Rhythmicity in Entorhinal Stellate Cells of the Juvenile Rat J Neurophysiol, December 1, 2008; 100(6): 3144 - 3157. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Gu, H. Hu, K. Vervaeke, and J. F. Storm SK (KCa2) Channels Do Not Control Somatic Excitability in CA1 Pyramidal Neurons But Can Be Activated by Dendritic Excitatory Synapses and Regulate Their Impact J Neurophysiol, November 1, 2008; 100(5): 2589 - 2604. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. D. Glasgow and C. A. Chapman Conductances Mediating Intrinsic Theta-Frequency Membrane Potential Oscillations in Layer II Parasubicular Neurons J Neurophysiol, November 1, 2008; 100(5): 2746 - 2756. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. M. Giocomo and M. E. Hasselmo Time Constants of h Current in Layer II Stellate Cells Differ along the Dorsal to Ventral Axis of Medial Entorhinal Cortex J. Neurosci., September 17, 2008; 28(38): 9414 - 9425. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
TOP
ABSTRACT
INTRODUCTION
