| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Experimental and Computational Neurophysiology Unit, Servicio Histología, Hospital Ramón y Cajal, Madrid, Spain
Submitted 6 July 2004; accepted in final form 26 September 2004
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
; Reyes 2001). Less is known on their role on the forward transmission of inputs. They are involved on the reshaping of the excitatory and inhibitory postsynaptic potentials (EPSPs/IPSPs), and the initiation of local spikes (Herreras 1990; Lipowsky et al. 1996; Magee and Johnston 1995; Martina et al. 2000; Stuart and Sakmann 1995; Williams and Stuart 2003). The apical shaft collects all apical inputs and may thus modulate and convey to the axon initial segment (AIS) an integrated signal separated from basal inputs. Whether local spikes on secondary branches will propagate or not into and beyond the apical shaft depends on a number of factors including its geometry, active properties, and specific local inhibition. Recently, we showed that this structure has a suitable geometry for safe forward conduction of APs by amplifying currents in the somatopetal direction (López-Aguado et al. 2002). We addressed here the participation of V-dependent currents in the apical shaft of CA1 pyramidal cells and inhibition on output decision.
The relation between synaptic, slow (subthreshold) and fast (spike-like) active dendritic events becomes highly complex in cells with multiple AP triggering zones (e.g., Larkum et al. 2001). In CA1 pyramidal cells, active dendritic currents/potentials have been shown either as slow or subthreshold components activated/recruited by synaptic inputs (Herreras 1990; Masukawa and Prince 1984; Turner et al. 1989) or as local or fully propagated APs (Andreasen and Nedergaard 1996; Fujita and Sakata 1962; Golding and Spruston 1998; Herreras 1990; Turner et al. 1991). The physiological meaning of the interaction between synaptic and V-dependent currents depends on whether the later contributes to the initiation or merely conduction of APs and whether these reach the axon or not. Using current source density analysis in vivo, we reported earlier a sequential activation of slow subthreshold and fast spike-like inward currents in proximal apical dendrites that were spatially segregated from their activating synaptic currents following activation of Schaffer collaterals (Herreras 1990; see also Kloosterman et al. 2001; Vida et al. 1995). We later reported that selective abolition of the subthreshold proximal currents caused by repeated spreading depression completely prevented population spikes (PSs) following maximal activation (Herreras and Somjen 1993). Since this result was obtained in semipathological conditions, we tried here to answer whether synaptic currents alone may initiate an AP (and where) without the participation of the en passant recruited apical currents in a more physiological situation. Also, because the alteration in local circuitry and the suppression of ongoing activity in the in vitro preparation might alter physiology, we made a comparative study in vitro and in vivo and also studied the modulatory effects of global and local inhibition. To this purpose, we blocked sodium channels by TTX microejection into the specific subcellular loci where AP initiation is bound to cell output: the AIS or the apical shaft. Evoked potentials and intrasomatic and intradendritic recordings were used.
The results indicate that synaptically recruited apical shaft currents constitute a major component of depolarizing potentials recorded in somata and are an absolute requirement to fire outgoing APs following synchronous Schaffer activation, whether initiated in the apical shaft or in the AIS. Apical shaft targeted inhibition appears enough to stop maximal Schaffer activation from firing APs in absence of recruited apical currents. Modulation of V-dependent currents in the apical shaft by intrinsic factors or inhibition may thus decide when and where an AP will be fired, performing as a variable filter for distal inputs. We also confirm previous findings indicating that APs may initiate in the apical shaft in vivo even for moderate excitatory inputs.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
Parasagittal hippocampal slices from the dorsal region (450 µm thick) were obtained from 120- to 150-g female Sprague-Dawley rats using a Vibroslice and placed in a holding chamber (35°C) at the interface between humidified air (95% O2-5% CO2) and artificial cerebrospinal fluid (ACSF) with the following composition (in mM): 120 NaCl, 3.0 KCl, 1.2 CaCl2, 1.2 MgSO4, 0.4 KH2PO4, 25 NaHCO3, and 10 glucose (pH 7.4). Slices were incubated for 
90 min and transferred to a recording chamber ("Oslo" type), continuously superfused by ACSF at 2 ml/min. Monopolar stimulating electrodes (40-µm tungsten wires) were placed at the alvear region and the lower third of the stratium radiatum. Three recording pipettes filled with 150 mM NaCl (3–6 M
) were placed along the main pyramidal cell axis to record from the same population. Two were located at the lower and proximal s. radiatum to record the fEPSP and dendritic PS, respectively (Herreras 1990; Turner et al. 1989). A third pipette was located either at the s. pyramidale or within the proximal s. oriens, where AISs ascend to the alveus. That the ortho- and antidromic-stimulating electrodes were activating essentially the same population was assessed by a collision test. Suitable stimulating/recording arrangements were chosen so that the antidromic PS was totally collided by a previous orthodromic one (Fig. 1A, inset). After filtering (1 Hz–5 kHz) and amplification, signals were recorded on VCR, acquired to a computer (20–40 kHz acquisition rate, Digidata 1200, Axon Instruments, Burlingame, CA), processed by Axotape and Axoscope software (Axon Instruments), and further analyzed by the Axum program (Trimetrix, Seattle, WA).
|
In some slices, an orthogonal cut was made across the CA1, sparing the lower third of the s. radiatum and the s. lacunosum. To rule out direct synaptic excitation of proximal loci, orthodromic stimulation was made in the fimbrial side of the cut, while recording and TTX ejection were made at the subicular side.
In a group of experiments, inhibition was blocked by adding 20 µM picrotoxin (PTX, Sigma) to the ACSF.
Intracellular recordings were obtained with pipettes backfilled with 4 M potassium acetate (60–120 M). Signals were amplified using a bridge circuit amplifier (Axoclamp 2B), filtered at 10 kHz, and stored on VCR for later analysis. Impalements were made from somata (Rin = 34.1 ± 0.9 M; n = 16) and apical dendrites of pyramidal cells (26 ± 2.4 M, n = 5). We considered healthy cells as those having a resting membrane potential of at least –65 mV.
In vivo experiments
Female Sprague-Dawley rats (200–250 g) were anesthetized with urethane (1.2–1.5 g/kg, ip) and fastened to a stereotaxic device. Surgical and stereotaxic procedures were as previously described (Herreras 1990). Two concentric bipolar stimulating electrodes were positioned in the alveus and in the ipsilateral CA3 for antidromic and Schaffer orthodromic activation of the CA1 pyramidal population, respectively (0.07- to 0.1-ms square pulses, 0.1–0.5 mA). A subcutaneous Ag/AgCl wire electrode under the neck skin was used as reference. The recording electrodes were connected to DC-coupled FET input stages. The characteristic configuration of evoked potentials guided the placement of the recording electrodes (Herreras 1990). Data processing was as in vitro.
Vertical arrays of two to three pipettes glued together were used to record from different depths at the same coordinates. Different barrels of the pipette assemblies were used to eject TTX (20 µM) and bicuculline methiodide (BIC, 100 µM) to cause local block of Na+ currents or remove GABAA-type inhibition, respectively. In some experiments, one pipette was filled with the glutamate receptor blocker 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX, 100 µM) to cause local block of excitatory transmission to assess the identity of the neuron population recorded at different layers. In the two-pipette assemblies, the intertip distance was 60 µm, except when indicated. In the three-pipette assemblies, the lower tip was at 300 and 360 µm from the upper ones to record from the s. pyramidale and proximal and lower s. radiatum, respectively. In our hands, solutions of PTX prepared as for in vitro experiments had no effect when microejected in vivo; thus we used BIC, which behaved as expected for GABAA blockade.
Measurement and statistics
The antidromic PS in the s. pyramidale was measured as the amplitude to the peak of its negative going limb with respect to the precedent baseline. When the artifact partially overlapped the evoked potential, the most positive value between them was used instead. In the s. radiatum, it was measured as the voltage from the negative peak (when discernible) to an imaginary line linking the precedent positive crest and the following inflection point in the rising phase. The orthodromic PS in the s. pyramidale was measured as the amplitude from the precedent positive crest and the negative peak, and in the s. radiatum, as the negative value to the peak with respect to the precedent baseline. This measurement contains a considerable component corresponding to the slower underlying envelope on which the dendritic PS rides on. The fEPSP was measured as the maximum slope of its negative-going phase and the fiber volley (FV) as the amplitude of the initial negative limb. The amplitude of the unitary AP was measured as the 90% value of the rising limb from rest and the rate of rise as the maximum slope. The AP threshold was measured as the voltage reached at the accelerating inflection noted at the base of their rising phase.
Data were quantified as the mean ± SE. Statistical comparisons were made using the Student t-test and considered significant when P < 0.05.
In this study, we have combined evoked potentials with intracellular recordings when possible. An advantage of field potentials is that they reflect average transmembrane currents in a population of unclamped cells, which cannot be achieved by unitary studies. Our later experimental and computational studies on the subcellular and macroscopic factors contributing to the PS have totally renewed the interpretation of this popular index and its changes, such as how the obligatory and varying interactions between different cell subregions modify the spatiotemporal map of the PS (López-Aguado et al. 2000–2002; Varona et al. 2000). Although laminar field analysis is the most reliable method, the use of multiple recordings along the different subregions of the same population may be sufficient for the current purposes because it provides safe information as to the causal and temporal relations between the different subcellular events of interest. The use of evoked potentials provides also a safe method to compare in vivo and in vitro results, because they are easily obtained in the two preparations. On the other hand, the intracellular study provides specific information on local processing within dendritic trees and details of the electrogenic repertoire that cannot be gathered from field potentials. The two recording methods are thus complementary.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
We first studied the effects of Na+ channel blockade in the main subcellular loci related to AP generation and propagation: the AIS and the apical shaft. Figure 1A shows a representative experiment and pooled data of the ortho- and antidromic field responses simultaneously recorded along different loci of the pyramidal cell axis, before and after TTX microejection from one of the pipettes (black vs. gray tracings, respectively). When TTX was ejected within the proximal s. radiatum to block Na+ currents in the apical shafts (Fig. 1A1), the FV (*), the fEPSP (arrow), and the somatic antidromic PS remained intact (99.5 ± 0.7, 99.8 ± 2.6, and 92.7 ± 5.2% of control, respectively), showing the local action of the drug. However, the orthodromic PSs were nearly abolished at the soma layer and strongly depressed in the proximal radiatum (13.1 ± 4.3 and 34.3 ± 4.2% of control, respectively), and backpropagated (antidromic) dendritic PSs were greatly reduced (42.9 ± 1.6%, n = 7). This result was also observed with double recordings in every trial on 34 slices (15 animals) in additional experiments, and indicates the essential participation of proximal Na+ currents on the initiation and forward conduction of APs and the poor ability of Schaffer-evoked synaptic currents alone to fire cells, not even at the AIS that remained fully functional. This interpretation was supported by the effects of local TTX within the proximal s. oriens (Fig. 1A2), where AISs ascend to the alveus. In this case, the antidromic but not the orthodromic somatic PS was greatly reduced (46.4 ± 6.4% of control). As expected, the somatic and apical PS remained unchanged, while backpropagation into basal dendrites reduced moderately (64.8 ± 4.6% of control; Fig. 1A2; n = 3).
Stronger local PS reduction can be readily achieved by increasing the ejected volume so TTX reached cells further away. However, we chose to restrict the spatial spread for the shake of clarity, since nearby dendritic loci began to be affected as well, which may contaminate the results. In fact, a possible explanation for the maximal Schaffer activation failing to initiate PSs could be that TTX may have blocked proximal afferent input activated by the bulk stimulating electrode that could go unnoticed by a distal recording electrode, lessening the total excitatory input. To check this possibility, we made a transversal cut across CA1 layers, sparing only the lower third of the s. radiatum, where the stimulating electrode was located (Fig. 1B). By recording at the opposite side of the cut, the ejection of TTX within the proximal apical region yielded identical results, i.e., the fEPSP and the antidromic somatic PS were unaffected while both dendrosomatic (ortho) and backpropagating (anti) PSs were greatly reduced (Fig. 1Bb; n = 3 slices).
To rule out possible misinterpretations due to the averaging nature of field potentials, intracellular somatic and apical dendritic recordings were performed. The unitary results confirmed the macroscopic findings in 11 somatic and 5 dendritic impalements. Thus TTX blockade of proximal apical Na+ currents abolished Schaffer-evoked but not antidromic APs in somata (Fig. 2A, top). On the contrary, when ejected at the AIS/soma region, the synaptically evoked APs recorded either at the main apical shaft (
100 µm from somata; Fig 2A, bottom; n = 3) or distal dendrites (300 µm away; Fig. 3A1; n = 2) remained, while backpropagating APs disappeared.
|
|
Proximal apical currents may decide the site of initiation of APs
The behavior of dendritic APs could be studied during the blockade of Na+ currents within the soma/proximal apical shaft. Intradendritic impalements were made 150 µm below the s. pyramidale (n = 5). Because of the slow recovery of the TTX effects, we limited the study to supramaximal and threshold activation. In controls (Fig. 3A1, black tracings), supramaximal Schaffer activation produced very short latency APs arising directly from the rising phase of the large depolarizing envelope. TTX ejected at the soma did not cause any change except for a slight increase of the later (arrow in blue tracings). When TTX was ejected repeatedly so to allow its spread further down within the apical shaft, a gradual decline of the AP amplitude was observed without a change on its latency or on the initial depolarization (Fig. 3A2), indicating a more distal dendritic site of AP initiation. Composite spikes were transiently observed (inset). Before the AP was totally blocked, a decrease of the stimulus intensity to a previously calculated threshold value yielded dendritic APs at a longer latency that departed from a smaller voltage than the maximum of the local EPSP. These spikes maintained the reduced amplitude (Fig. 3A3). A few shocks later, the AP blocked completely without development of partial spikes, indicating that the AP initial site in control was at the AIS for lower activation intensities. TTX ejected at the soma or proximal dendrites never modified the intradendritic EPSP recorded more distally (Fig. 3, A and B).
Ejection of TTX within the proximal apical shaft of a distally impaled dendrite yielded a characteristic sequence before the expected abolition of the AP (Fig. 3B). The short-latency AP first decreased without a change of latency and then split into a stationary decreasing spike and a delaying AP that maintained a rather constant amplitude (Fig. 3Ba, curved arrows). At threshold intensity (Fig. 3Bb), the longer latency AP also decreased its amplitude to the same extent before complete blockade. A similar decrease occurred for the antidromic AP (data not shown). These results are compatible with a transient shift on the AP initiation site from the apical dendrite to the AIS before the total extinction by effect of proximal TTX. Somatic impalements showed small spikelets following TTX on their proximal apical shafts (Fig. 3C, arrows), likely corresponding to distally initiated APs that failed to regenerate across the TTX-treated zone.
Somatic depolarization is strongly contributed by apical recruited currents on activation of distant synapses
The contribution of recruited apical currents to somatic depolarization following distant synaptic activation was studied by delivering TTX into the proximal apical dendrites (Fig. 4). The somatic resting membrane potential was not affected by apical TTX (66.7 ± 1.0 vs. 66.5 ± 0.9 mV, respectively; n = 9). We found a notable decrease of the somatic depolarizing envelope regardless of the stimulus intensity. Responses that were subthreshold before TTX decreased to 46% afterward (measured as the maximum peak value; n = 4; Fig. 4A). The fEPSP slope simultaneously measured in the s. radiatum remained unchanged. At suprathreshold intensity (Fig. 4B), the maximum value could not be measured in controls, and the AP threshold was used instead. This was 11.0 ± 1.6 mV, while the maximum value of the depolarizing envelope following apical TTX was 6.3 ± 2.6 mV (range, 2.6–8.7 mV; n = 5). During high intensity, some cells transiently fired at a delayed latency after proximal apical TTX. In these cases, the AP threshold also decreased from 11.4 ± 4.2 to 9.4 ± 2.7 mV, as expected for a shift on the AP initiation locus as seen from the soma.
|
Removal of inhibition renders far synaptic currents able to fire APs in vitro without apical boost
The unsuccessful initiation of APs by Schaffer excitation when proximal Na+ currents are blocked may be due to the premature shunt of synaptic currents by inhibition on their way to trigger zones. We checked this possibility by adding the GABAA channel blocker PTX to the bath (50 µM). After 15 min of PTX perfusion, multiple PSs had developed (Fig. 5A1, red vs. black tracings), a slow negative component became patent in the proximal s. radiatum (arrow in 2), and the fEPSP enlarged without a change on its initial phase (3). When TTX was ejected at the proximal s. radiatum (pipette 2) on disinhibited slices (blue vs. red tracings), the fEPSP and the antidromic somatic PS still remained unaffected, but the orthodromic apical PS was blocked, as expected. However, the orthodromic PS at the soma layer was not abolished. Instead, it delayed, decreased, and widened, indicating a less synchronous but still powerful pyramidal firing. Figure 5B shows the quantification of this result by selecting experiments in which TTX caused a 50% reduction of the PS in the s. radiatum. In these experiments, apical TTX caused a reduction of the somatic PS to 34.5 ± 4.3% in control and to 75.7 ± 6.1% during PTX perfusion (n = 7). The longer latency of the PS during PTX + TTX indicated a shift of the AP initial site to a more distant locus (the AIS). This result was confirmed with intracellular recordings (Fig. 5C). In the presence of PTX, apical TTX still caused a decrease of the somatic depolarizing envelope (see time plots at fixed instants a and c), while the apparent threshold for the AP (b) increased (arrows). The same result was observed in n = 3 cells.
|
Because of the possibility that dendritic excitability was different in vitro than in vivo, we first measured the PS amplitude at the soma and proximal apical layers (100 µm below) obtained at two times the threshold intensity. After correction for increased tissue resistivity in vitro (37% higher than in vivo; López-Aguado et al. 2002), the orthodromic PS was 10.6 ± 0.3 mV in the soma layer and 6.9 ± 0.5 mV in proximal s. radiatum (n = 21), while in vivo measured 11.3 ± 0.8 mV in the soma layer and 14.1 ± 0.9 mV in the proximal s. radiatum (n = 8). That is, in vitro the apical PS is 47% smaller than its soma counterpart, while in vivo it is 20% larger. In part, the excess amplitude of the apical PS in vivo was due to a more conspicuous underlying slow negative envelope (Fig. 6A, small arrow).
|
Despite the above relative amplitude difference between somatic and proximal PS in vivo and in vitro, maximal Schaffer activation behaved similarly. Thus TTX ejection at the AIS locus in the proximal s. oriens decreased the antidromic but not the orthodromic PS, while basal backpropagation was reduced (Fig. 7, A1 and B1). Lowering the pipette assembly so to eject TTX into the soma layer decreased further the antidromic PS (60%) and its backpropagation and began to affect also the orthodromic somatic PS (<30%), but its proximal apical counterpart was unaffected (Fig. 7, A2 and B2). Finally, TTX ejection into the proximal radiatum decreased the local PS and orthodromic somatic PS to <30%, while the antidromic PS was unaffected (Fig. 7, A3 and B3). Similar qualitative results were obtained in two, four, and six TTX ejections at the AIS, soma, and proximal dendritic layers, respectively, in five additional animals.
|
The effect of local removal of inhibition was studied by microejection of the GABAA blocker BIC (see METHODS) from one pipette after blockade of apical currents with TTX ejected from another pipette (blue and red in the scheme of Fig. 8A correspond to BIC and TTX filled pipettes, respectively). After TTX was ejected in the proximal apical shaft, the orthodromic PS began to decrease in both apical and soma layers as expected (blue tracings and arrow in Fig. 8). Shortly thereafter, BIC ejection at the soma layer caused the fast recovery of the somatic PS, while the apical one recovered normally (red tracings and arrow in Fig. 8). The soma antidromic PS remained unchanged in all cases (plots on the right in Fig. 8A and open filled triangles in Fig. 8B; compare with Fig. 7B3). Notably, the latency difference between somatic and apical PSs (soma delayed) that was unaffected by TTX ejection, reversed after BIC, indicating that the initiation of cell firing shifted to the AIS (Fig. 8B, top). Similar qualitative results were obtained in three animals.
|
Since the strength of synaptic excitation has been claimed an important factor modulating the site of AP initiation in these cells (Golding and Spruston 1998), we checked in three animals the effects of TTX on threshold PSs (<30% of maximal amplitude) by adjusting the stimulus intensity. A representative experiment is shown in Fig. 9. Strong ongoing variability of the PS amplitude was customary (Fig. 9A). The first observation was that the dendritic PS still had a shorter latency than the somatic one (0.4 ms, Fig. 9B, top), even if the absolute latency jitter between trials was notable (>0.4 ms). When TTX was ejected at the proximal s. oriens (arrow) aiming the blockade of AIS Na+ currents, we observed the expected decrease of the antidromic somatic PS and its backpropagation (Fig. 9B, circles). However, the Schaffer-evoked small PSs were unaffected (triangles). Only a small reduction of the delay between dendritic and somatic PSs was noticed (top) that can be interpreted as a decrease of the net somatic current contribution to the local PS as it became increasingly dominated by volume propagated dendritic currents (López-Aguado et al. 2000; Varona et al. 2000). These results indicate that, even for moderate Schaffer excitatory input, the APs are also initiated at the apical shaft in a sizable portion of the population, although some may not reach the soma.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
The presence of V-dependent currents in pyramidal dendrites is well known, and evidence for a direct impact on cell output is rapidly accumulating (see references in the Introduction). They are of greatest relevance in the proximal apical shaft, because of its strategic location taking up the majority of inputs from the apical tree (Bannister and Larkman 1995; Trommald et al. 1995). Proximal apical potentials/currents can be functionally discriminated from the synaptic ones, so they appear as all-or-nothing or stepwise subthreshold unitary events (Masukawa and Prince 1984; Regher et al. 1993; see also Fig. 3) or as spatially segregated macroscopic potentials or sinks of current (Herreras 1990; Herreras and Somjen 1993; Turner et al. 1989). Pharmacological tools (TTX, GABA blockers) may also help discriminate between synaptic and recruited V-dependent currents/potentials, as shown here. The correlation of unitary and population active dendritic events is not always straightforward, particularly as it concerns to subthreshold events and distal spike activity. For instance, it remains to be determined whether the proximal nonsynaptic envelope-like sink of current that we described earlier following Schaffer activation (Herreras 1990) and that appeared instrumental for AP firing (Herreras and Somjen 1993) can be ascribed to the recruited V-dependent currents found in whole cell studies. In this study, the results obtained with field potentials and intracellular recordings in vitro are in perfect agreement. This was expected because we used mostly supramaximal intensities that override the ongoing activity of different neurons. The gross relations described here can be extended in future work by studying the dynamic features of the envelope-like currents using current source density analysis in combination with pharmacological manipulations in vivo. It would also be most important to expand on the details of initiation and cooperation among distal local spikes and their relation to APs in the apical shaft and AIS, which would be better accomplished by multiple recordings on thin secondary dendrites and the main apical shaft on a single neuron.
In this work, it was clear that the slower subthreshold proximal potentials are more conspicuous in vivo than in vitro and appeared more resistant to TTX blockade (cf. Figs. 7A3 and 8A2). It is reasonable to think that the electrogenic mechanism for these currents involves both Na+ and Ca2+ channels and is not fully recovered from the preparation procedures. A considerably smaller dendritic PS in vitro also points to this interpretation.
Apical currents have been involved on EPSP carving (Lipowsky et al. 1996), long-term plasticity (Taube and Schwartzkroin 1988), and cross-talk between different spike-generating zones (Larkum et al. 2001; Magee and Carruth 1999). Their relation to cell output is more controversial. At present, the prevalent idea derived mostly from whole cell studies in vitro is that apical currents merely facilitate backpropagation of APs initiated at the AIS, while apical initiation and full forward propagation occur rarely and is bound to unusual large excitatory input (e.g., Golding and Spruston 1998). The results shown here modifies and extends on this notion since maximal Schaffer activation is poorly effective to initiate APs, either at the apical shaft or at the low threshold AIS, unless apical currents are recruited on the way. These recruited currents appear as a necessary amplifier to decide cell output on intense, synchronous, and spatially concentrated synaptic input, and highlight the critical gating role of the apical shaft: a large synaptic input and local boosting may not be sufficient.
In our study, TTX did not affect the AISs or somata when ejected in the proximal region, as shown by the permanence of full-size antidromic PSs and unitary APs. A possible reduction of the total excitatory input by TTX spread was also ruled out by surgically limiting the input to a narrow distal bridge and by the invariance of distal unitary EPSPs and fEPSPs after TTX ejection to the soma or proximal apical layers. The identity of the population of neurons recorded with the electrodes placed at different strata in vivo and in vitro was also confirmed by the expected effect (or lack of effect) of the different drugs applied locally on the evoked activity of other cell subregions (e.g., Fig. 6).
In a way, these results question the extended idea that CA1 apical dendrites have a low excitability because they rarely initiate and conduct APs forwardly. It can be argued that excitatory input is poorly synchronized on individual cells during ongoing activity and will rarely reach enough voltage as to activate Na+ channels. While this awaits direct experimental assessment in vivo, there is no precise quantification of the number of synchronous synapses required to fire a pyramidal cell. Our computer simulations predict that an excitatory conductance equivalent to as few as 15–20 synchronous excitatory synapses may suffice to initiate cell firing (López-Aguado et al. 2002; see also Mel 1999). We estimate from data from Bannister and Larkman (1995) that a single CA1 pyramid may receive from 1,000 to 3,000 Schaffer contacts in the distal third of the s. radiatum. The number of simultaneously active inputs may vary enormously, and the degree of synchronization will decide whether a dendritic spike will be initiated, conducted in a continuous or pseudosaltatory way to the axon or simply backpropagated from there. The necessary degree of excitatory synchronization may be largely surpassed during some physiological electrographic patterns, as hippocampal sharp waves (e.g., Kamondi et al. 1998). In any case, beyond the role of dendritic spikes, these results underscore the importance of subthreshold envelope-like currents. These currents may play quite a different role than spikes and serve the purpose of modulating/deciding cell output under different input patterns.
We found that small PSs evoked by reduced Schaffer excitation are also initiated at the apical shaft in vivo (see also Kloosterman et al. 2001). The reduction of latency difference between dendritic and somatic PSs is, however, indicative of partial failure of active forward propagation; hence some individual APs, although initiated in dendrites, may not reach the soma/axon. We already arrived to the same conclusion in our original current source density study (Herreras 1990). It is not easy to predict whether isolated local spikes generated on distal dendritic branches would reach the soma/axon without the upholding effect of recruited V-dependent currents in the apical shaft (e.g., Vetter et al. 2001), simply because these are obligatorily recruited. Imaging techniques and multiple recording experiments on a single cell may help solving this question. On the other hand, we showed that apical currents are less conspicuous in vitro than in vivo and therefore the failure of local spikes to propagate forwardly may have been overestimated. One possibility is that dendritic electrogenic machinery may not recover fully from the in vitro preparation procedures (see Kirov et al. 2004). A lower excitability of apical dendrites would rise the threshold for apical AP initiation that would be biased toward the AIS. Another possibility is the strong modulation of dendritic Na+ channels by metabolic and external factors (e.g., Gasparini and Magee 2002) that might not be properly maintained between physiological ranges during, for instance, whole cell recordings. The relevant result is that maximal Schaffer activation, which presumably initiated a considerable number of local spikes (e.g., Fig. 3), is poorly efficient at firing APs not even at the AIS without the assistance of apical currents.
An important issue arises: how much of the somato-axonic depolarization is indeed of synaptic origin and how much corresponds to nonsynaptic current recruited on the way? Our intracellular recordings showed that at least for Schaffer activation about one-half of the EPSP amplitude at the soma is originated by current recruited in the apical shaft. In neocortical pyramidal cells, this contribution appears to be restricted to the somato-axonic region (Stuart and Sakmann 1995; but see Larkum et al. 2001). One may consider that the apical contribution is negligible for isolated inputs. Thus synaptic currents scale with distance to the soma in CA1 pyramidal cells (Magee and Cook 2000), a mechanism that may serve to maintain the somato-axonic impact of EPSPs regardless of the input site. In this case, the apical shaft would behave largely as a passive cable. However, this and earlier works indicate that this function is overridden when a certain number of inputs summate and start recruiting V-dependent currents. For instance, local spikes (Golding and Spruston 1998; Kamondy et al. 1998; Masukawa and Prince 1984; Regher et al. 1993; this report) are a well-established electrogenic event whose integrative rules shall prevail over those of the elementary synaptic inputs from which they were born out. The debate on sub- or supralinear summation of distal inputs at the somato-axonic triggering area (e.g., Cash and Yuste 1999; Polsky et al. 2004) should be translated to the nearest location at the distal end of the apical shaft because the present experimental results and previous computational studies (López-Aguado et al. 2002) do indicate that, once the AP is initiated in this structure, cell output is ensured.
Since the different parts of pyramidal cells are subject to specific inhibitory control (e.g., Miles et al. 1996; Pouille and Scanziani 2004; Williams and Stuart 2003), it can be presumed that the final impact of specific inputs will be largely dependent on the functional state of local circuits in a behaviorally controlled manner. It is well known that, during behaviors associated to hippocampal theta rhythm inhibition increases in CA1, decreasing/blocking Schaffer evoked PSs (Herreras et al. 1986, 1987; Leung and Vanderwolf 1980), a process in which both pre- and postsynaptic inhibition participate (Herreras et al. 1988). Here we used bath-applied PTX in vitro and local BIC in vivo to mimic a general decrease of inhibition or a specific disinhibition of the apical shaft in vivo, respectively. In both cases, Schaffer-evoked synaptic currents "restored" their efficiency and succeeded in firing APs and PSs after apical blockade of Na+ currents, although the initial site shifted to the AIS. The initial AP site would be decided from a subtle equilibrium between local AP threshold, the electrotonic distance of synaptic inputs to all trigger zones, their number and synchronism, and the level of inhibition (e.g., Shen et al. 1999). Since inhibition is far less synchronized during ongoing activity than following "bulk" electrical stimulation, it is not unreasonable to think that a few far located excitatory synaptic currents alone might initiate firing during normal brain operation. Ongoing modulation of inhibition may thus be the major factor to decide not just cell output but also the AP initial site (e.g., Chen et al. 1997; Kim et al. 1995; Staff and Spruston 2003; Tsubokawa and Ross 1996), acting both as a logical gate for specific inputs and as a switch between global operating modes of apical dendrites. When distal spikes do not propagate across, inhibition would be acting as a variable shunt regulating the amount of current transferred to the soma/axon. Inhibition may also turn on and off the output-deciding capabilities of the apical shaft and hence whether the concurrence of basal inputs will be required or not. Taking into account that the apical tree receives inputs from different brain regions, it appears advantageous a mechanism of output decision that may be operated under particular combinations of inputs. These results might indicate, for instance, that at least two different inputs must coincide to cause cell output or, on the contrary, a single type of input might suffice, depending on the degree of global or input-specific inhibition and the intrinsic modulation of V-dependent currents.
We envisage the role of the apical shaft as a multiplexing device. On one side, it may secure the transfer of synaptic inputs with high fidelity to the soma/AIS region, where the output decision will be made by weighing all inputs from the apical and basal trees, but it may also decide cell output by itself. The two functions (loss-free transmission and boosting) are compatible and may constitute an efficient computational device to switch between global functioning modes of pyramidal cells.
|
GRANTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
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 |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
Bannister NJ and Larkman AU. Dendritic morphology of CA1 pyramidal neurones from the rat hippocampus. II. Spine distributions. J Comp Neurol 360: 161–171, 1995.[CrossRef][Web of Science][Medline]
Cash S and Yuste R. Linear summation of excitatory inputs by CA1 pyramidal neurons. Neuron 22: 383–394, 1999.[CrossRef][Web of Science][Medline]
Chen WR, Mitgaard M, and Shepherd GM. Forward and backward propagation of dendritic impulses and their synaptic control in mitral cells. Science 278: 463–468, 1997.
Fujita Y and Sakata H. Electrophysiological properties of CA1 and CA2 apical dendrites of rabbit hippocampus. J Neurophysiol 25: 209–222, 1962.
Gasparini S and Magee JC. Phosphorylation-dependent differences in the activation properties of distal and proximal dendritic Na+ channels in rat CA1 hippocampal neurons. J Physiol 541: 665–72, 2002.
Golding NL and Spruston N. Dendritic sodium spikes are variable triggers of axonal action potentials in hippocampal CA1 pyramidal neurons. Neuron 21: 1189–1200, 1998.[CrossRef][Web of Science][Medline]
Herreras O. Propagating dendritic action potential mediates synaptic transmission in CA1 pyramidal cells in situ. J Neurophysiol 64: 1429–1441, 1990.
Herreras O, Solís JM, Herranz AS, Martín del Río R, and Lerma J. Sensory modulation of hippocampal transmission. II. Evidence for a cholinergic locus of inhibition in the Schaffer-CA1 synapse. Brain Res 451: 303–313, 1988.
Herreras O, Solís JM, and Lerma J. Abolition of CA1 population spike by sensory stimulation. Exp Brain Res 61: 654–657, 1986.[Web of Science][Medline]
Herreras O, Solís JM, Martín del Río R, and Lerma J. Characteristics of CA1 activation through the hippocampal trisynaptic pathway in the unanaesthetized rat. Brain Res 413: 75–86, 1987.[CrossRef][Web of Science][Medline]
Herreras O and Somjen GG. Effects of prolonged elevation of potassium on hippocampus of anesthetized rats. Brain Res 617: 194–204, 1993.[CrossRef][Web of Science][Medline]
Johnston D, Magee JC, Colbert CM, and Christie BR. Active properties of neuronal dendrites. Ann Rev Neurosci 19: 165–186, 1996.[CrossRef][Web of Science][Medline]
Kamondi A, Acsady L, and Buzsaki G. Dentritic spikes are enhanced by cooperative network activity in the intact hippocampus. J Neurosci 18: 3919–3928, 1998.
Kim HG, Beierlein M, and Connors BW. Inhibitory control of excitable dendrites in the neocortex. J Neurophysiol 74: 1810–1814, 1995.
Kirov SA, Petrak LJ, Fiala JC, and Harris KM. Dendritic spines disappear with chilling but proliferate excessively upon rewarming of mature hippocampus. Neuroscience 127: 69–90, 2004.[CrossRef][Web of Science][Medline]
Kloosterman F, Peloquin P, and Leung L. Apical and basal orthodromic population spikes in hippocampal CA1 in vivo show different origins and patterns of propagation. J Neurophysiol 86: 2435–2444, 2001.
Larkum ME, Zhu JJ, and Sakmann B. Dendritic mechanisms underlying the coupling of the dendritic with the axonal action potential initiation zone of adult rat layer 5 pyramidal neurons. J Physiol 533. 2: 447–466, 2001.
Leung LS and Vanderwolf CH. Behavior-dependent evoked potentials in the hippocampal CA1 region of the rat. II. Effect of eserine, atropine, ether and pentobarbital. Brain Res 198: 119–133, 1980.[CrossRef][Web of Science][Medline]
Lipowsky R, Gillessen T, and Alzheimer C. Dentritic Na+ channels amplify EPSPs in hippocampal CA1 pyramidal cells. J Neurophysiol 76: 2181–2191, 1996.
López-Aguado L, Ibarz JM, and Herreras O. Modulation of dendritic action currents decreases the reliability of population spikes. J Neurophysiol 83: 1108–1114, 2000.
López-Aguado L, Ibarz JM, and Herreras O. Activity-dependent changes of tissue resistivity in the CA1 region in vivo are layer-specific: modulation of evoked potentials. Neuroscience 108: 249–262, 2001.[CrossRef][Web of Science][Medline]
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.
Magee JV and 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 and Cook EP. Somatic EPSP amplitude is independent of synapse location in hippocampal pyramidal neurons. Nat Neurosci 3: 895–903, 2000.[CrossRef][Web of Science][Medline]
Magee JC and Johnston D. Synaptic activation of voltage-gated channels in the dendrites of hippocampal pyramidal neurons. Science 268: 301–304, 1995.
Martina M, Vida I, and Jonas P. Distal initiation and active propagation of action potentials in interneuron dendrites. Science 287: 295–300, 2000.
Masukawa LM and Prince DA. Synaptic control of excitability in isolated dendrites of hippocampal neurons. J Neurosci 4: 217–227, 1984.[Abstract]
Mel B. Why have dendrites? A computational perspective. In: Dendrites, edited by Stuart G, Spruston N, and Häusser M. New York: Oxford University Press, 1999, p. 271–289.
Miles R, Toth K, Gulyas AI, Hajos N, and Freund TF. Differences between somatic and dendritic inhibition in the hippocampus. Neuron 16: 815–823, 1996.[CrossRef][Web of Science][Medline]
Polsky A, Mel BW, and Schiller J. Computational subunits in thin dendrites of pyramidal cells. Nat Neurosci 7: 621–627, 2004.[CrossRef][Web of Science][Medline]
Pouille F and Scanziani M. Routing of spike series by dynamic circuits in the hippocampus. Nature 429: 717–723, 2004.[CrossRef][Medline]
Regher W, Kehoe J, Ascher P, and Armstrong C. Synaptically triggered action potentials in dendrites. Neuron 11: 145–151, 1993.[CrossRef][Web of Science][Medline]
Reyes A. Influence of dendritic conductances on the input-output properties of neurons. Annu Rev Neurosci 24: 653–75, 2001.[CrossRef][Web of Science][Medline]
Shen GY, Chen WR, Midtgaard J, Shepherd GM, and Hines ML. Computational analysis of action potential initiation in mitral cell soma and dendrites based on dual patch recordings. J Neurophysiol 82: 3006–3020, 1999.
Staff NP and Spruston N. Intracellular correlate of EPSP-spike potentiation in CA1 pyramidal neurons is controlled by GABAergic modulation. Hippocampus 13: 801–805, 2003.[CrossRef][Web of Science][Medline]
Stuart G and Sakmann B. Amplification of EPSPs by axosomatic sodium channels in neocortical pyramidal neurons. Neuron 15: 1065–76, 1995.[CrossRef][Web of Science][Medline]
Taube JS and Schwartzkroin PA. Mechanisms of long-term potentiation: a current-source density analysis. J Neurosci 8: 1645–1655, 1988.[Abstract]
Trommald M, Jensen V, and Andersen P. Analysis of dendritic spines in rat CA1 pyramidal cells intracellularly filled with a fluorescent dye. J Comp Neurol 353: 260–274, 1995.[CrossRef][Web of Science][Medline]
Tsubokawa H and Ross WN. IPSPs modulate spike backpropagation and associated [Ca2+]i changes in the dendrites of hippocampal CA1 pyramidal neurons. J Neurophysiol 76: 2896–2906, 1996.
Turner RW, Meyers DE, and Barker JL. Localization of tetrodotoxin-sensitive field potentials of CA1 pyramidal cells in the rat hippocampus. J Neurophysiol 62: 1375–87, 1989.
Turner RW, Meyers DER, Richardson TL, and Barker JL. The site for initiation of action potential discharge over the somatodendritic axis of rat hippocampal CA1 pyramidal neurons. J Neurosci 11: 2270–2280, 1991.[Abstract]
Varona P, Ibarz JM, López-Aguado L, and Herreras O. Macroscopic and subcellular factors shaping CA1 population spikes. J Neurophysiol 83: 2192–2208, 2000.
Vetter PH, Roth A, and Haüsser M. Propagation of action potentials in dendrites depends on dendritic morphology. J Neurophysiol 85: 926–937, 2001.
Vida I, Czopf J, and Czéh G. A current-source density analysis of the long-term potentiation in the hippocampus. Brain Res 671: 1–11, 1995.[CrossRef][Web of Science][Medline]
Williams SR and Stuart GJ. Voltage- and site-dependent control of the somatic impact of dendritic IPSPs. J Neurosci 23: 7358–7367, 2003.
This article has been cited by other articles:
|
|
 |
|
  S. Gasparini and J. C. Magee State-Dependent Dendritic Computation in Hippocampal CA1 Pyramidal Neurons J. Neurosci., February 15, 2006; 26(7): 2088 - 2100. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Canals, I. Makarova, L. Lopez-Aguado, C. Largo, J. M. Ibarz, and O. Herreras Longitudinal Depolarization Gradients Along the Somatodendritic Axis of CA1 Pyramidal Cells: A Novel Feature of Spreading Depression J Neurophysiol, August 1, 2005; 94(2): 943 - 951. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
), but not the synaptically evoked PS (
). Basal backpropagation reduced in the 2 cases (
and 
). When TTX was ejected at the soma (2), the antidromic (
t > 0, top plot in B). Thus apical initiation is maintained even for reduced inputs. Note the strong variability of the PS amplitude. After basal TTX, the antidromic PS decreased (circles) but not the orthodromic PSs (triangles), which maintained similar variability and amplitude. A slight reduction of the latency difference between the apical and the somatic PS was observed that may indicate a partial failure of apical APs to invade the soma (see RESULTS).