| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
TRANSLATIONAL PHYSIOLOGY
1Experimental and Computational Neurophysiology Laboratory, Department of Investigación-Histología, Hospital Ramón y Cajal, Madrid; 2Department of Matemática Aplicada, School of Biology, University Complutense of Madrid, Madrid; and 3School of Cieucias Experimentales y Salud, University of San Pablo, Madrid, Spain
Submitted 8 November 2004; accepted in final form 23 March 2005
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
) is a propagating mass event that runs slowly through nervous tissue associated to abrupt ion changes in the extracellular space and large negative DC potential (Vo). In spite of a wealth of phenomenological data, the biophysical nature of SD is poorly understood (reviewed in Bure
et al. 1974; Marshall 1959; Somjen 2001). Both glia and neurons depolarize during SD (Sugaya et al. 1975), but neurons appear to be the main active players (e.g., Herreras et al. 1994; Largo et al. 1996, 1997a,c ). The large negative Vo signal is considered an extracellular index of cell depolarization, but its spatiotemporal distribution has not been conclusively correlated to specific cells or parts of cells. Neither is it known whether cell depolarization is passive to transmembrane ion redistribution or caused by active changes of membrane conductance in individual cells or both.
Neurons become silent during SD passage and recordings from individual cells show near complete depolarization, extreme shunt of membrane input resistance (Rin), and loss of electrical responsiveness (e.g., Herreras and Somjen 1993c; Morlock et al. 1964; Snow et al. 1983; Sugaya et al. 1975). Different lines of evidence point to dendrites as a major place for SD (Herreras and Somjen 1993a; Ochs and Hunt 1960), but their precise role is uncertain because available recordings are only from cell somata. The inactivated electrical status of somata has been presumed for the entire somatodendritic membranes, although direct confirmation is missing. It is unknown whether the depolarization is homogeneous throughout the entire neuron morphology or how it relates to the extracellular Vo or the massive ion changes. The relevance of this issue is that according to core conductor theory if neurons are uniformly depolarized across their entire anatomy, they will not generate transmembrane current across their membranes, hence they would most likely play a passive role in the SD-associated electrical display.
We earlier described sharp Vo and interstitial ion gradients during SD along the pyramidal cell axis in the hippocampal CA1 (Herreras and Somjen 1993b) that suggest spatially heterogeneous events along the morphology of individual pyramidal neurons. Regional differences in light scattering (Aitken et al. 1998; Basarski et al. 1998) point in the same direction. Also additional evidence argues against an exclusive role of increased extracellular potassium in causing neuronal depolarization. Thus the sustained depolarization recorded in cell somata is at odds with the rapidly decreasing potassium during the time course of SD (Herreras and Somjen 1993b) or the decreasing holding (inward) current in voltage-clamped single cells during SD (Czéh et al. 1993).
To bring into harmony the macroscopic and unitary results, we performed a subcellular study of membrane excitability during SD in the CA1 region, the architectonical simplicity of which allows a fair correlation of intracellular and field potentials at the subcellular level. We combined intrasomatic and intradendritic recordings with DC and evoked field potentials to test local membrane potential (Vm), input resistance (Rin), and excitability in vitro and in vivo. The results showed a stereotyped spatiotemporal pattern of local dendritic Vm and Rin changes along the longitudinal axis of pyramidal cells during SD that cannot be detected from somatic impalements. The SD-associated subcellular changes may run independently in the basal and apical dendritic trees. Strong depolarization gradients arose along the main longitudinal axis of individual cells that necessarily involve source-sink transmembrane current distribution. In close agreement, subregions far from the SD-affected membranes remained excitable and may still develop evoked activity. We suggest that an increase of membrane conductance in specific dendritic subregions of individual neurons is on the origin of neuron depolarization and the macroscopic changes related to SD. Some results have been presented in abstract form (Largo et al. 1997b).
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Hippocampal slices were prepared from female Sprague-Dawley rats (120–150 g) using standard techniques. Briefly, the animal was anesthetized with ether and decapitated and the brain was removed to chilled artificial cerebrospinal fluid (ACSF). Parasagittal slices from the dorsal region (450 µm thick) were obtained using a Vibroslice and transferred to a holding chamber (32°C) at the interface between humidified air (95% O2-5% CO2) and oxygenated ACSF with the following composition (in mmol/l): 120 NaCl, 3.0 KCl, 1.8 CaCl2, 1.8 MgSO4, 0.4 KH2PO4, 20 NaHCO3, and 11 glucose (pH 7.4). Slices were incubated for 
90 min and transferred for recording to an interface Oslo chamber superfused by ACSF at 2 ml/min (32°C). Anti- and orthodromic activation were achieved by stimuli (0.04 ms) delivered through monopolar electrodes (40-µm tungsten wires) placed in the alvear region and in the strata (s.) oriens, radiatum, or lacunosum-moleculare (lac-mol), respectively. The Ag/AgCl wires of recording pipettes were connected to DC-coupled input stages. After filtering (DC-5 kHz) and amplification, signals were stored on a VCR, and processed off-line by the computer programs Axotape (Axon Instruments, Burlingame, CA) and Axum (Trimetrix, Seattle, WA). Acquisition was made at a sampling rate of either 5–20 Hz for slow signals or 20 to 40 kHz for evoked potentials (Digidata 1200, Axon).
Evoked field responses were studied during SD waves the associated negative DC signal of which affected partially or entirely the pyramidal cell morphology. Somatodendritic population spikes (PSs, ortho- and antidromically initiated), field excitatory postsynaptic potentials (fEPSPs), and other slow synaptic and nonsynaptic potentials (see Herreras 1990; Herreras and Somjen 1993c) were scrutinized in different positions along the main cell axis.
In vivo experiments
Experiments were performed in the dorsal CA1 region of urethan-anesthetized (1.2 g/kg ip) female Sprague-Dawley rats (200–250 g) maintained at 37 ± 0.1°C. Surgery and stereotaxic procedures were as previously described (Herreras 1990; Largo et al. 1996). Concentric bipolar stimulating electrodes were placed in the alveus for antidromic activation and in the ipsilateral CA3 or the s. lac-mol to activate Schaffer or perforant path fibers, respectively. Recording electrodes were glass micropipettes filled with isotonic Krebs-Ringer solution (5–10 M
) and placed at the s. pyramidal or at different depths within dendritic layers guided by the typical evoked potentials (Herreras 1990; Varona et al. 2000). Pipettes were sometimes used in vertical arrays of two to three, glued to record from different depths (Herreras and Somjen 1993b). Recording and signal analysis was made as in vitro.
Intracellular study
Micropipettes (1.5 mm OD) were backfilled with 4 M potassium acetate (80–100 M). Signals were amplified using a bridge circuit amplifier (Axoclamp 2A, Axon Instruments), filtered at 10 kHz, and stored in VCR or acquired into a computer (20–40 kHz, acquisition rate). Impalements were made from somata (in vivo and in vitro) and apical dendrites (in vitro) of pyramidal cells. The impalements in vitro were obtained 80 µm below the slice surface, and another pipette located extracellularly within <50 µm of the impaled element was used to calculate true transmembrane potential (Vm) by off-line subtraction. In our experience, healthy cells have resting membrane potential more negative than –63 mV (–66.3 ± 2.7 mV, n = 7) and APs overshooting by 30 mV at the soma. Dendritic impalements were obtained mainly within the s. radiatum and typically displayed larger EPSPs and smaller APs. The average Rin calculated from the linear range of I-V plots in somata in vivo was 30.2 ± 4.2 M (range, 22–41), slightly higher in vitro (35 ± 3.2 M, range: 28–44). Dendritic recordings had a smaller Rin (25.1 ± 1,9 M, n = 14). Two additional dendrites had 52 and 70 M. We found no apparent differences of intrinsic excitability or evoked synaptic response related to Rin.
Elicitation of SD waves
SD waves were elicited by pressure ejection of potassium solutions (0.2–1.2 M KCl, high-K+) from a glass micropipette that was also used for recording. Pressure was adjusted to release the minimum volume necessary to elicit a single SD wave (typically microdrops 
50 nl). Multiple SDs could be elicited using this procedure without apparent damage to the tissue. The ejecting pipette was located in the basal or apical dendritic arbors, 100 µm above or below the pyramidal cell layer, respectively. Pressure was applied either manually, through an air-filled syringe or by means of a Picospritzer (General Valve). SD waves were always initiated 500 µm away from the intracellular pipette to ensure that a true propagating event was being studied. Local nonpropagating events were sometimes generated that could be easily recognized as they were only detected at the local injecting extracellular pipette.
Local values of Vm and Rin were obtained for the time course of SD in soma and apical dendritic impalements, whereas excitability was measured by synaptic activation of different afferent bundles along the dorsoventral extension. Typically, SD affects both the basal and the apical trees simultaneously in CA1. However, the basal and apical SD components may appear separately, behaving as independent basal or apical SD waves (Herreras and Somjen 1993a–d). We were able to control the sequence of the strata invaded by SD. Thus high-K+ microinjections within the s. oriens elicited SD waves that traveled only through this stratum (termed here basal-only SDs) or that preceded a yoked apical wave (basal-first SDs). Apical microinjections elicited the more standard apical-first SD waves in which the apical front led the basal front by 2–4 s (Herreras and Somjen 1993a). Basal-only SDs were easily elicited and usually extended also to the s. pyramidal, whereas apical-only SDs were rarely generated by this procedure. Therefore the small leading interval in apical-first SDs was used to collect somatodendritic measurements during "apical-only" SD periods.
The experiments followed EC regulations on animal care and handling, and every effort was taken to avoid suffering of the animals during experimentation.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Extracellularly recorded standard SD waves in CA1 showed a characteristic spatial and temporal profile defined by the area of tissue undergoing a strong negative Vo (Herreras and Somjen 1993a,b). Initially, the SD covered all but the most distal portion of the basal and apical dendritic trees. The basal component terminated earlier and the apical wave became gradually restricted to a narrow dendritic band 150–200 µm wide centered around 200 µm below the s. pyramidal. Typically, the wave front led in the s. radiatum by 2–4 s over the s. pyramidal and oriens. These regional differences are much more pronounced in vivo. Intra- and extracellular voltage changes during SD are time correlated but change in opposite directions after the initial abrupt shift. Whereas Vi decayed slowly from the initial maximum at any loci, the corresponding Vo became increasingly negative in equivalent quantity, resulting in a quasi-steady near-zero Vm (Fig. 1A). This was a constant finding at any subcellular domain exhibiting a large negative Vo (note the matching times of Vm and local Vo marked by the arrows for somatic and apical dendritic recordings in Fig. 1B, 1 and 2, respectively). However, because the Vo signal lasted longer within the s. radiatum than in the soma layer and basal dendrites (Fig. 1B), different cell domains remained near zero for different periods, creating longitudinal gradients of Vm depolarization within individual cells. Typically, the Vm at pyramidal somata may have recovered to half its initial depolarization by the end of the wave (arrowhead in Fig. 1B) while it remained fully depolarized at the proximal apical dendritic locus for the duration of the extracellular local SD. In vitro, the soma Vm recovery was smaller by the end of the apical wave (70.3 ± 6.4 of the initial maximum 
Vm, n = 6, measured at the instant when the apical SD wave began to recover).
|
Vo from the Vi for three different SD periods in which the negative Vo extended across well-defined bands along the pyramidal cell main axis, depending on the locus of initiation (see METHODS). These were a basal SD (encroaching the s. oriens and usually but not always also the s. pyramidal), an apical SD (spanning the s. radiatum) and a mixed SD (encroaching both subregions). Because SD could be forced to lead in one or another strata, the initial seconds were used to define basal- and apical-only SD periods. Figure 2A shows a group of basal-first SDs. The end of the basal-only SD period is marked by the arrows, when the apical front arrived to the electrode location. During basal-only SD periods, a soma-to-apical depolarization gradient was found from –34.5 ± 1.0 mV (Vi 30 mV, n = 49) at the soma to –55 mV (Vi 10 mV) within the most distally recorded apical dendrite (370 µm away, Fig. 2B, open circles). However, when SD arrived to the apical tree, a constant maximum depolarization to about –20 mV (Vi = 43.8 ± 0.8 mV, n = 5) was measured throughout the apical dendrites impaled anywhere in the s. radiatum (black circles in Fig. 2, A and B). In apical-first SDs, the arrival of the basal front caused no further increase in the local apical Vm as compared with apical-only periods (see Fig. 3A2). Presumably, a sizable Vm gradient would also be present in apical SDs from the thicker dendrites to the distal apical tuft (see following text). These values correspond to the intracellular voltage shift (Vi), while Vm required substantial correction by the local Vo (see the summarized results in Table 1). Thus during apical SD, (Vi = 43.8 ± 0.8 mV, n = 5) the local Vm depolarized 63.6 mV from rest (local Vo = –19.8 ± 2.0 mV, n = 3). Thus Vo is a major participant in local Vm during SD. Besides, because Vi longitudinal depolarization gradients are smoother than those of Vo that showed sharp dorsoventral limits, the Vm differences between near cell subregions at the fringe of the extracellular SD are stronger than could be inferred solely from Vi measurements. For instance, during a basal-only SD period, a small positive local Vo in the apical tree (3.2 ± 0.3 mV, n = 39) decreased somewhat the net local Vm, whereas during apical SD, the strong negative Vo boosted the effective apical depolarization 
35 mV in different times and locations (double-headed arrows in Fig. 2A). We were able to accurately measure the longitudinal instant depolarization gradient from the soma to the proximal apical shaft during basal-only SD. Thus at the time of the maximum Vi, the somata had depolarized by 52.4 mV on average while the proximal apical shafts did so only by 20 mV (data pooled within the closer 100 µm; Table 1). Notably, if the somatic depolarization was measured against a ground reference, the absolute values of Vi would reflect poorly the actual extension of dendritic membranes affected by SD (see groups 1, 3, and 5 in Table 1). Globally, these results indicate the existence of intracellular strong and quasi-steady potential gradients from fully depolarized membranes to SD-spared regions, the negative Vo being a macroscopic marker of the fully depolarized affected membranes. Because uneven depolarization promotes axial currents that must close loop across the membrane, we next explored the local properties of membrane resistance.
|
|
|
Classic work with sharp electrodes described an almost full breakdown of membrane resistance during SD. The question again is whether the entire cell morphology undergoes such a "breakdown." This issue is essential as in that case neurons could hardly raise any substantial net current. Because available recordings had been made only from cell somata, we checked here the evolution of subcellular conductance by measuring Rin along the soma and apical dendrite membranes by applying short hyperpolarizing current pulses (50 ms, –0.5 nA). Larger evoked EPSPs, smaller APs, and a strong amplitude reduction in the successive APs during a depolarizing current pulse (1 s, 0.4 nA) characterized distal dendrites (Fig. 3A1). At the outset of a standard apical-first SD, Rin fell to undetectable values (<5% of control) at any recorded intracellular loci between the soma (n = 16) and apical dendrites up to the mid s. radiatum (n = 5). Curiously, Rin began to recover slowly at any loci shortly after the initial peak (Fig. 3B). Right before the Vm initiated the fast recovery, the Rin had regained 34.9 ± 1.6% of the control value (range: 7.4–73.8, n = 70). Full recovery took place at a faster rate in parallel to Vm and the negative Vo at the end of the wave. However, when the SD affected only the basal tree (basal-only periods), the initial Rin dropped much less and in a gradient-like fashion, from undetectable values at the soma and proximal apical shaft to 70% of control within the more distally recorded dendrites (arrows in Fig. 3A2, basal first, and plot 1 in Fig. 3B). Although we could not check Rin within the thin distal dendrites, the trend along the apical tree suggests that Rin drop would be negligible. In standard apical/basal SD waves, the partial recovery phase of Rin during SD always proceeded at a faster rate the further from the SD-active regions in the soma and proximal dendrites (see 3 successive instants in Fig. 3B) suggesting a somatopetal wave-like shut down of membrane conductance. Thus the membrane shunt is largest at the cell domains with the most negative Vo, while a decreasing gradient of local Rin was observed at distant loci lacking negative Vo (e.g., apical dendrites during basal only-SD).
Loss of electrical excitability occurs only at subcellular domains matching the negative Vo
The existence of stationary depolarizing gradients opens the possibility that SD-spared subcellular regions remain, at least in part, electrically excitable. We first investigated in vivo whether the entire somatodendritic membrane of CA1 pyramidal cells is electrically unexcitable during a standard SD wave moving across both dendritic arbors by recording evoked field potentials at different somatodendritic loci. Antidromic, Schaffer-evoked and s. oriens-evoked orthodromic activity were completely abolished for the entire duration of the longer negative Vo in the s. radiatum (Fig. 4A1), consistent with the idea of a complete loss of membrane excitability along the entire neuron morphology (64 trials in 16 animals). However, we extended the analysis to the previously unchecked afferent band in the distal most apical tree, i.e., the s. lac-mol. In agreement with the preceding prediction, the fEPSP evoked and recorded from this dendritic band, where SD negativity had very low-amplitude, was reduced, but not abolished (Fig. 4A2; n = 6 trials in 3 animals). The sparing of distal evoked activity was better explored in vitro by using multiple recording electrodes within the band of interest (Fig. 4B). Again, the fEPSP evoked and recorded within the s. lac-mol was barely affected (c), while that recorded slightly more proximal was abolished (b). Potentials evoked and/or recorded anywhere in the s. radiatum were totally abolished (a; n = 6 slices). These results indicate that during SD some parts of the same neurons are inactivated whereas others are not.
|
). The typical negative anti- and orthodromic PSs vanished within the SD-affected somato-basal strip (dashed area Fig. 5A, inset). Instead, an orthodromic field consisting of positive slow and spike components developed that closely mirrored those in the s. radiatum (Fig. 5B, arrows). This indicated that the SD-active subcellular domain constituted a preferential membrane locus for the return of far active currents to the extracellular space.
|
; Herreras 1990).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Neurons are not electrically inactivated during SD
For decades, there has been the tacit assumption that neurons were electrically impaired during SD. This concept arose from two long known facts: neurons become electrically unresponsive, and potassium floods the ES. The idea of a global membrane breakdown and failure of ionic homeostatic mechanisms prevailed, a chaotic scenario in which all cells were thought depolarized and neuron membranes leaky. The present results refute such notion as far as it concerns to the electrical state of neurons and question the idea of a breakdown of ionic homeostasis.
All previous studies have shown that neuronal somatic Vm approaches zero during SD (Collewijn and Van Harreveld 1966; Higashida et al. 1974; Snow et al. 1983) except in rare cases when some cells appeared not to "fall" into SD (e.g., Czéh et al. 1993; Sugaya et al. 1975). Also, in the multilayered cortex and retina several unitary and ensemble activities had been reported in layers out of the mainstream of SD (Muñoz 1970; Rosenblueth and García-Ramos 1966; Tomita 1984). In these cases, it was postulated that some dendrites within the SD-affected area could be passively depolarized by increased potassium released form actively SD-engaged cell elements, resulting in increased excitability. The present results in the monolayered CA1 demonstrate that some parts of all pyramidal cells actively engaged in SD may still be electroresponsive even if the distant somata cannot detect such activity. Thus all intracellular recordings from soma showed a profound drop of the SD-associated Rin, while distal-most dendritic evoked activity remained intact. When SD was forced to run only through the basal tree, this result was even more evident, as the entire apical tree was able to generate almost normal activity. The failure of somatic recordings to detect distant activity resides on two facts: first, it takes place at an electrically remote place and second, the somata and proximal dendrite membrane themselves are fully depolarized and shunted, avoiding any distant current to reach the somatic electrode. The electrical compartmentalization of neuron subregions is even stronger during SD: shunted membranes isolate electrically the SD-spared membranes from the rest of the cell. The fact that SD is a highly local membrane event that may occur in discrete zones of one or another dendritic tree, with or without the soma, whereas the adjacent membrane regions are only partly depolarized, accounts for both the common observation of "silenced" cells and the rare cases of partial unitary depolarization in the hippocampus. The general principles underlying these observations make us suggest that they may well apply to any other brain region undergoing SD. The sharp longitudinal depolarizing gradient established along the main pyramidal axis is therefore compatible with normal electrical properties of farther well polarized subcellular regions, which may then act as current sources for depolarized regions and still generate individual or collective (field) electrogenic responses.
Longitudinal depolarization gradients, shunted membranes and loops of current
The nature of the near-zero depolarization reached during SD has not been clarified so far. Early attempts to use the volume conductor theory proposed the axons as the current source for a generalized somatodendritic sink and assumed a global depolarization of the entire membrane (Monakhov et al. 1962). This idea would be hardly tenable because homogeneous depolarization and membrane shunt would not generate any transmembrane current. Besides, axons could not drain a substantial amount of current because of their small size and the high resistance of the axon-soma junction (López-Aguado et al. 2002). Unfortunately, the spatial resolution of that study was very poor and could not detect regional heterogeneity of the SD-associated Vo (Herreras and Somjen 1993a,b,d). We now report that inactivating depolarization occurs only in a discrete band of SD-affected dendritic membranes, closely matching the extracellular negative band and the ion changes. As expected during a steady and profound regional decrease of Rm, most outward current should occur through the shunted membranes themselves and to a lesser extent through the adjacent regions that display a decreasing gradient of depolarization. An earlier study proposed inward current across dendrite membranes to account for the Vo profile of SD in the cerebellum (Nicholson and Kraig 1981), which is in agreement with our results. Our results indicate the sparing of the distal most section in both dendritic trees, what is a necessary requirement to achieve a sizable Im. We have not attempted to impale the thin distal dendrites, but we did show evoked activity in the s. lac-mol, demonstrating that the dendrites in the apical tuft are spared during a standard SD. Therefore they must keep Vm and Rm near resting values and remain excitable. A detailed CSD analysis of the Vo and resistivity profiles in vivo is required to adequately harmonize the present subcellular data and the macroscopic potential and ion changes.
At least three possible sources of depolarization have to be considered. First, the reduction of concentration gradients of the permeant ions Na+ and K+ shifted their respective equilibrium potentials closer to zero. We calculated that the experimentally measured ion ranges would take the new SD "resting" Vm below –20 mV. However, it is important to note that because extracellular ion variations are also restricted to specific bands of dendrites (Herreras and Somjen 1993b), the local "resting" Vm would also vary along the neuron anatomy, giving rise to axial currents and transmembrane current loops.
A second mechanism of depolarization is the opening of channels the combined net current of which will discharge the membrane capacitance to a depolarizing level. Because the conductance is extremely large and sustained, the intra- and extracellular ion concentrations could reasonably reach their equilibrium potential, hence, little current should flow on this account. A third source of depolarization results from the spatially restricted activation of channels in specific dendritic bands. As for classic synaptic potentials, current flows internally from the more depolarized regions away, enters from the extracellular space across the SD-active zone and leaves out through less depolarized regions. Uneven depolarization is necessary to originate axial and net transmembrane currents. The magnitude of the axial currents is directly proportional to the Vm difference between adjacent regions. Therefore the sharpness of the internal gradient limits the maximum inward current, hence the magnitude of Im. The net Im will cause a negative potential in the outer parts of the depolarized membrane, a mechanism that is common to the generation of for instance the customary negative fEPSPs during synaptic activation (Herreras 1990). Thus a major part of the near-zero Vm depolarization (40 mV) is actually given by the large negative Vo resulting from the net inward current.
The combination of strong zonal depolarization and intense membrane shunt create quasi-steady longitudinal depolarization gradients along the main axis of pyramidal cells and intense transmembrane current loops. Ultimately, the steady depolarization is maintained by the decrease of transmembrane ion gradients that could be promoted by intense current flowing across standard V-dependent channels at the initial moments of SD. While this may be so, the subsequent evolution (i.e., that of the global SD reaction) will depend on which channel types remain open. Using our detailed biophysical model of the CA1 pyramidal cell (Varona et al. 2000) to clamp Vm near zero we found that kinetics allowed none in the standard battery of V-dependent channels to remain open long enough as to produce substantial inward current. Besides, a number of observations indicate that inward current during SD, and the negative Vo are not mediated through standard V-dependent channels (Czéh et al. 1993; Jing et al. 1993; Müller and Somjen 2000). Altogether, this let us postulate that an additional channel independent of voltage opens to account for the extreme zonal loss of membrane resistance. Based on calculations of the pore size, Kraig and Nicholson (1978) proposed a large SD-specific membrane channel. The candidate channels should be able to sustain prolonged currents so to account for the long and variable SD duration under different circumstances (Herreras and Somjen 1993a,b,d) without being restrained by the strict kinetics of most V-dependent inward channels. However, the present results indicate that whatever channel is open during SD will be used to let current in driven by the internal longitudinal depolarization gradients. The identity of the channel is relevant from the point of view of clinical applications but not for the understanding of the ultimate biophysical mechanism of SD.
A notable observation is that Vm remained steady in SD-active domains while Vo and Vi varied in opposite directions. One possibility is that Vo and Vi must balance to each other because Vm is clamped to a fixed value defined by the local transmembrane ion gradients and the large local increase in conductance. Any change of the Vo would force an equivalent opposite change on Vi and vice versa. Among the factors that may affect one or another during the time course of SD are the varying spatial cancellation of sinks and sources during the narrowing of SD-affected dendritic membranes, and variations of the local tissue resistivity. This view is consistent with the temporal evolution of Rin. Thus the slow recovery of Rin at the soma (Fig. 3B) would reflect the closure of channels in locations progressively closer to the soma, where the intracellular pipette is injecting the current pulses. This pattern is reminiscent of some wave-like intracellular process shutting down SD channels. Thus the smooth somatopetal decrease of Rin away from SD active regions is not necessarily caused by a decreasing gradient of open channels. It can be well explained by the large space-clamp error of sharp pipettes. Such interpretation is consistent with the smaller decrease of Rm found in whole cell recordings at the soma (Czéh et al. 1993) as compared with sharp electrode measurements.
We envisage the global process of SD as a succession of membrane events mediated by different channels extending along partially overlapping subcellular domains, each yielding the necessary conditions for the next to open. During SD, neurons remain electrically functional except for the "anomalous" behavior that represents the opening of a normally closed conductance in specific subregions of their anatomy. The presence or absence of this conductance on cell types might be the cause of the selective vulnerability of different neurons and brain regions.
|
GRANTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Present address of S. Canals: Max Planck Institute for Biological Cybernetics 72076, Tübingen, Germany.
|
FOOTNOTES |
|---|
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: O. Herreras, Dept. Investigación-Histología, Hospital Ramón y Cajal, Ctra. Colmenar km 9, Madrid 28034, Spain (E-mail: oscar.herreras{at}hrc.es)
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Basarski TA, Duffy SN, Andrew RD, and MacVicar BA. Imaging spreading depression and associated intracellular calcium waves in brain slices. J Neurosci 18: 7189–7199, 1998.
Bure J, Bureová O, and Krivánek J. The Mechanism and Application of Leão's Spreading Depression of Electroencephalographic Activity. Prague: Academia, 1974.
Canals S, López-Aguado L, and Herreras O. Synaptically-recruited apical currents are required to initiate axonal and apical spikes in hippocampal pyramidal cells: modulation by inhibition. J Neurophysiol 93: 909–918, 2005.
Collewijn H and Van Harreveld A. Membrane potential of cerebral cortical cells during spreading depression and asphyxia. Exp Neurol 15: 425–436, 1966.[CrossRef][ISI][Medline]
Czéh G, Aitken PG, and Somjen GG. Membrane currents in CA1 pyramidal cells during spreading depression (SD) and SD-like hypoxic depolarization. Brain Res 632: 195–208, 1993.[CrossRef][ISI][Medline]
Herreras O. Propagating dendritic action potential mediates synaptic transmission in CA1 pyramidal cells in situ. J Neurophysiol 64: 1429–1441, 1990.
Herreras O and Somjen GG. Propagation of spreading depression among dendrites and somata of the same cell population. Brain Res 610: 276–282, 1993a.[CrossRef][ISI][Medline]
Herreras O and Somjen GG. Analysis of potentials shifts associated with recurrent spreading depression and prolonged unstable SD induced by microdialysis of elevated K+ in hippocampus of anesthetized rats. Brain Res 610: 283–294, 1993b.[CrossRef][ISI][Medline]
Herreras O and Somjen GG. Effects of prolonged elevation of potassium on hippocampus of anesthetized rats. Brain Res 617: 194–204, 1993c.[CrossRef][ISI][Medline]
Herreras O and Somjen GG. Prolonged unstable depression: a modified manifestation of spreading depression in rat hippocampus. In: Slow activity Changes of the Brain, edited by Haschke W, Speckmann EJ, and Roitbak AI. Boston: Birkhäuser, 1993d, p. 129–138.
Herreras O, Largo C, Ibarz JM, Somjen GG, and Marín del Río R. Role of neuronal synchronizing mechanisms in the propagation of spreading depression in the in vivo hippocampus. J Neurosci 14: 7087–7098, 1994.[Abstract]
Higashida H, Mitarai G, and Watanabe S. A comparative study of membrane potential changes in neurons and neuroglial cells during spreading depression in the rabbit. Brain Res 65: 411–425, 1974.[CrossRef][ISI][Medline]
Jing J, Aitken PA, and Somjen GG. Role of calcium channels in spreading depression in rat hippocampal slices. Brain Res 604: 251–259, 1993.[CrossRef][ISI][Medline]
Kraig RP and Nicholson C. Extracellular ionic variations during spreading depression. Neuroscience 3: 1045–1059, 1978.[CrossRef][ISI][Medline]
Largo C, Cuevas P, Somjen GG, Martín del Río R, and Herreras O. The effect of depressing glial function on rat brain in situ on ion homeostasis, synaptic transmission and neuronal survival. J Neurosci 16:1219–1229, 1996.
Largo C, Ibarz JM, and Herreras O. Effects of the gliotoxin fluorocitrate on spreading depression and glial membrane potential in rat brain in situ. J Neurophysiol 78: 295–307, 1997a.
Largo C, Ibarz JM, and Herreras O. Inactivation of Cell Soma Does Not Preclude Apical Shaft Currents in CA1 Neurons. 27th Annual Meeting Soc. Neurosci. New Orleans, LA, 1997b.
Largo C, Tombaugh G, Aitken PG, Herreras O, and Somjen GG. Heptanol but not fluoroacetate prevents the propagation of spreading depression in rat hippocampal slices. J Neurophysiol 77: 9–16, 1997c.
Leão AAP. Spreading depression of activity in the cerebral cortex. J Neurophysiol 7: 359–390, 1944.
López-Aguado L, Ibarz JM, Varona P, and Herreras O. Structural inhomogeneities differentially modulate action currents and population spikes initiated in the axon or dendrites. J Neurophysiol 88: 2809–2820, 2002.
Marshall WH. Spreading cortical depression of Leão. Physiol Rev 39: 239–279, 1959.
Monakhov KK, Fifková E, and Bure J. Steady potential field of hippocampal spreading depression. J Cell Comp Physiol 59: 155–161, 1962.[Medline]
Morlock NL, Mori K, and Ward AA. A study of single cortical neurons during spreading depression. J Neurophysiol 27: 1192–1198, 1964.
Müller M and Somjen GG. Na+ and K+ concentrations, extra and intracellular voltages, and the effect of TTX on hypoxic rat hippocampal slices. J Neurophysiol 83: 735–745, 2000.
Muñoz E. Facilitation of cortical cell activity during spreading depression. J Neurobiol 2: 47–60, 1970.[Medline]
Nicholson C and Kraig RP. The behavior of extracellular ions during spreading depression. In: The Application of Ion-Selective Microelectrodes, edited by Zeuten T. Elsevier-North Holland, 1981, chapt. 13, p. 217–238.
Ochs S and Hunt K. Apical dendrites and propagation of spreading depression in cerebral cortex. J Neurophysiol 23: 432–444, 1960.
Rosenblueth A and García-Ramos J. Some phenomena usually associated with spreading depression. Acta Physiol Latinoam 16: 141–179, 1966.[Medline]
Snow RW, Taylor CP, and Dudek E. Electrophysiological and optical changes of rat hippocampus during spreading depression. J Neurophysiol 50: 561–572, 1983.
Somjen GG. Mechanisms of spreading depression and hypoxic spreading depression-like depolarization. Physiol Rev 81: 1065–1096, 2001.
Sugaya E, Takato M, and Noda Y. Neuronal and glial activity during spreading depression in cerebral cortex of cat. J Neurophysiol 38: 822–841, 1975.
Tomita T. Spreading depression potential (SDP) in the frog retina. An Acad Brasil Cien 56: 505–517, 1984.
Varona P, Ibarz JM, López-Aguado L, and Herreras O. Macroscopic and subcellular factors shaping CA1 population spikes. J Neurophysiol 83: 2192–2208, 2000.
This article has been cited by other articles:
|
|
 |
|
  J. Makarova, J. M. Ibarz, S. Canals, and O. Herreras A Steady-State Model of Spreading Depression Predicts the Importance of an Unknown Conductance in Specific Dendritic Domains Biophys. J., June 15, 2007; 92(12): 4216 - 4232. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 O. Herreras, G. Somjen, and A. Strong Electrical prodromals of spreading depression void Grafstein's potassium hypothesis J Neurophysiol, November 1, 2005; 94(5): 3656 - 3657. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
