Partial Spikes Research Articles

A novel spiking neural network of receptive field encoding with groups of neurons decision

Human information processing depends mainly on billions of neurons which constitute a complex neural network, and the information is transmitted in the form of neural spikes. In this paper, we propose a spiking neural network (SNN), named MD-SNN, with three key features: (1) using receptive field to encode spike trains from images; (2) randomly selecting partial spikes as inputs for each neuron to approach the absolute refractory period of the neuron; (3) using groups of neurons to make decisions. We test MD-SNN on the MNIST data set of handwritten digits, and results demonstrate that: (1) Different sizes of receptive fields influence classification results significantly. (2) Considering the neuronal refractory period in the SNN model, increasing the number of neurons in the learning layer could greatly reduce the training time, effectively reduce the probability of over-fitting, and improve the accuracy by 8.77%. (3) Compared with other SNN methods, MD-SNN achieves a better classification; compared with the convolution neural network, MD-SNN maintains flip and rotation invariance (the accuracy can remain at 90.44% on the test set), and it is more suitable for small sample learning (the accuracy can reach 80.15% for 1000 training samples, which is 7.8 times that of CNN).

Stability of cluster solutions in a cooperative consumer chain model

We study a cooperative consumer chain model which consists of one producer and two consumers. It is an extension of the Schnakenberg model suggested in Gierer and Meinhardt [Kybernetik (Berlin), 12:30-39, 1972] and Schnakenberg (J Theor Biol, 81:389-400, 1979) for which there is only one producer and one consumer. In this consumer chain model there is a middle component which plays a hybrid role: it acts both as consumer and as producer. It is assumed that the producer diffuses much faster than the first consumer and the first consumer much faster than the second consumer. The system also serves as a model for a sequence of irreversible autocatalytic reactions in a container which is in contact with a well-stirred reservoir. In the small diffusion limit we construct cluster solutions in an interval which have the following properties: The spatial profile of the third component is a spike. The profile for the middle component is that of two partial spikes connected by a thin transition layer. The first component in leading order is given by a Green's function. In this profile multiple scales are involved: The spikes for the middle component are on the small scale, the spike for the third on the very small scale, the width of the transition layer for the middle component is between the small and the very small scale. The first component acts on the large scale. To the best of our knowledge, this type of spiky pattern has never before been studied rigorously. It is shown that, if the feedrates are small enough, there exist two such patterns which differ by their amplitudes.We also study the stability properties of these cluster solutions. We use a rigorous analysis to investigate the linearized operator around cluster solutions which is based on nonlocal eigenvalue problems and rigorous asymptotic analysis. The following result is established: If the time-relaxation constants are small enough, one cluster solution is stable and the other one is unstable. The instability arises through large eigenvalues of order O1. Further, there are small eigenvalues of order o1 which do not cause any instabilities. Our approach requires some new ideas: (i) The analysis of the large eigenvalues of order O1 leads to a novel system of nonlocal eigenvalue problems with inhomogeneous Robin boundary conditions whose stability properties have been investigated rigorously. (ii) The analysis of the small eigenvalues of order o1 needs a careful study of the interaction of two small length scales and is based on a suitable inner/outer expansion with rigorous error analysis. It is found that the order of these small eigenvalues is given by the smallest diffusion constant ε2(2).

Slow integration leads to persistent action potential firing in distal axons of coupled interneurons

The conventional view of neurons is that synaptic inputs are integrated on a timescale of milliseconds to seconds in the dendrites, with action potential initiation occurring in the axon initial segment. In a subset of rodent hippocampal and neocortical interneurons in acute slices prepared from serotonin 5b receptor (Htr5b) BAC transgenic mice [1], we found a much slower form of integration leading to action potential initiation in the distal axon. In approximately 80% of these interneurons (n=214 of 274), and in 23% of hippocampal interneurons in wild-type C57BL/6 mice (n=6 of 26), hundreds of spikes, evoked over a period of minutes, resulted in persistent firing that lasted for a similar duration. Persistent firing was observed in response to step current injections, synaptic stimulation, sine wave current injections or in response to stimulation with natural spike trains [2]. With all of these protocols, multiple stimuli were required to induce persistent firing. While axonal action potential firing was required to trigger persistent firing, somatic depolarization was not; antidromic stimulation of the axon while hyperpolarizing the soma with current injection produced persistent firing. In addition, phase plots of persistent firing revealed that spikes had two components: an initial component represented spiking in the axon and a second component that overlapped with the current-evoked spikes, indicative of a somato-dendritic spike following an initial, axonally initiated spike. In some recordings (n = 11), partial spikes (spikelets) were observed during persistent firing. These spikelets overlapped the first component of the full-amplitude spikes, with the peak of the spikelets corresponding to an inflection on the rising phase seen in the full-amplitude spikes. These observations suggest that the first component of each action potential during persistent firing is an axonal spike, which sometimes fails to evoke a somato-dendritic spike. Furthermore, in some cells (n=3), spikelets were observed if the soma was hyperpolarized during persistent firing. These spikelets were smaller than those observed without hyperpolarization, suggesting that they are caused by propagation failures at a more distal axonal location. Using a stylized computational model constructed with the NEURON simulation environment [3] of a branching axon attached to a soma, we simulated both small- and large amplitude spikelets, as well as full-amplitude spikes, by depolarizing a branch of the axon during somatic hyperpolarization. Large-amplitude spikelets corresponded to failure of the action potential to invade the soma, whereas small-amplitude spikelets corresponded to failures at the axon branches, 40 μm from the soma. Similar results were obtained with a full morphological model of a branching axonal arborization. Additionally, in paired recordings, persistent firing was not restricted to the stimulated neuron; it could also be produced in the unstimulated cell (n=3). None of these pairs exhibited direct electrical coupling, and both glutamate and GABA receptors were blocked. Consolidating these results suggests the existence of a previously unknown operational mode for some mammalian neurons. These interneurons can slowly integrate spiking, share the output across a coupled network of axons and respond with persistent firing even in the absence of input to the soma or dendrites.

Connexin36 vs. connexin32, “miniature” neuronal gap junctions, and limited electrotonic coupling in rodent suprachiasmatic nucleus

Suprachiasmatic nucleus (SCN) neurons generate circadian rhythms, and these neurons normally exhibit loosely-synchronized action potentials. Although electrotonic coupling has long been proposed to mediate this neuronal synchrony, ultrastructural studies have failed to detect gap junctions between SCN neurons. Nevertheless, it has been proposed that neuronal gap junctions exist in the SCN; that they consist of connexin32 or, alternatively, connexin36; and that connexin36 knockout eliminates neuronal coupling between SCN neurons and disrupts circadian rhythms. We used confocal immunofluorescence microscopy and freeze-fracture replica immunogold labeling to examine the distributions of connexin30, connexin32, connexin36, and connexin43 in rat and mouse SCN and used whole-cell recordings to re-assess electrotonic and tracer coupling. Connexin32-immunofluorescent puncta were essentially absent in SCN but connexin36 was relatively abundant. Fifteen neuronal gap junctions were identified ultrastructurally, all of which contained connexin36 but not connexin32, whereas nearby oligodendrocyte gap junctions contained connexin32. In adult SCN, one neuronal gap junction was >600 connexons, whereas 75% were smaller than 50 connexons, which may be below the limit of detectability by fluorescence microscopy and thin-section electron microscopy. Whole-cell recordings in hypothalamic slices revealed tracer coupling with neurobiotin in <5% of SCN neurons, and paired recordings (>40 pairs) did not reveal obvious electrotonic coupling or synchronized action potentials, consistent with few neurons possessing large gap junctions. However, most neurons had partial spikes or spikelets (often <1 mV), which remained after QX-314 [ N-(2,6-dimethylphenylcarbamoylmethyl)triethylammonium bromide] had blocked sodium-mediated action potentials within the recorded neuron, consistent with spikelet transmission via small gap junctions. Thus, a few “miniature” gap junctions on most SCN neurons appear to mediate weak electrotonic coupling between limited numbers of neuron pairs, thus accounting for frequent detection of partial spikes and hypothetically providing the basis for “loose” electrical or metabolic synchronization of electrical activity commonly observed in SCN neuronal populations during circadian rhythms.

Layer‐specific pyramidal cell oscillations evoked by tetanic stimulation in the rat hippocampal area CA1 in vitro and in vivo

Tetanic stimulation of axons terminating in the CA1 region of the hippocampus induces oscillations in the gamma-to-beta frequency band (13-100 Hz) and can induce long-term potentiation (LTP). The rapid pyramidal cell discharge is driven by a mainly GABA(A)-receptor-mediated slow depolarization and entrained mainly through ephaptic interactions. This study tests whether cellular compartmentalization can explain how cells, despite severely reduced input resistance, can still fire briskly and have IPSPs superimposed on the slow GABAergic depolarization, and whether this behaviour occurs in vivo. Oscillations induced in CA1 in vitro by tetanic stimulation of the stratum radiatum or oriens were analysed using intracellular and multichannel field potentials along the cell axis. Layer-specific effects of focal application of bicuculline indicate that the GABAergic depolarization is concentrated on tetanized dendrites. Current-source density analysis and characteristics of partial spikes indicate that early action potentials are initiated in the proximal nontetanized dendrite but cannot invade the tetanized dendrite, where recurrent EPSPs and evoked IPSPs were largely suppressed. As the oscillation progresses, IPSPs recover and slow the neuronal firing to beta frequencies, with a small subpopulation of neurons continuing to fire at gamma frequency. Carbonic anhydrase dependence, threshold intensity, frequency, field strength and spike initiation/propagation of tetanus-evoked oscillations in urethane-anaesthetized rats, validate our observations in vitro, and show that these mechanisms operate in healthy tissue. However, the disrupted electrophysiology of the tetanized dendrites will disable normal information processing, has implications for LTP induction and is likely to play a role in pathological synchronization as found during epileptic discharges.

Synaptic and nonsynaptic contributions to giant ipsps and ectopic spikes induced by 4-aminopyridine in the hippocampus in vitro.

Hippocampal slices bathed in 4-aminopyridine (4-AP, < or =200 microM) exhibit 1) spontaneous large inhibitory postsynaptic potentials (IPSPs) in pyramidal cells, which occur without the necessity of fast glutamatergic receptors, and which hence are presumed to arise from coordinated firing in populations of interneurons; 2) spikes of variable amplitude, presumed to be of antidromic origin, in some pyramidal cells during the large IPSP; 3) bursts of action potentials in selected populations of interneurons, occurring independently of fast glutamatergic and of GABA(A) receptors. We have used neuron pairs, and a large network model (3,072 pyramidal cells, 384 interneurons), to examine how these phenomena might be inter-related. Network bursts in electrically coupled interneurons have previously been shown to be possible with dendritic gap junctions, when the dendrites were capable of spike initiation, and when action potentials could cross from cell to cell via gap junctions; recent experimental data showing that dendritic gap junctions between cortical interneurons lead to coupling potentials of only about 0.5 mV argue against this mechanism, however. We now show that axonal gap junctions between interneurons could also lead to network bursts; this concept is consistent with the occurrence of spikelets and partial spikes in at least some interneurons in 4-AP. In our model, spontaneous antidromic action potentials can induce spikelets and action potentials in principal cells during the large IPSP. The probability of observing this type of activity increases significantly when axonal gap junctions also exist between pyramidal cells. Sufficient antidromic activity in the model can lead to epileptiform bursts, independent of alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) and N-methyl-D-aspartate (NMDA) receptors, in some principal cells, preceded by IPSPs and spikelets. The model predicts that gap junction blockers should suppress large IPSPs observed in 4-AP and should also reduce the probability of observing antidromic activity, or bursting, in pyramidal cells. Experiments show that, indeed, the gap junction blocking compound carbenoxolone does suppress spontaneous large IPSCs, occurring in 4-AP plus ionotropic glutamate blockers, together with a GABA(B) receptor blocker; carbenoxolone also suppresses large, fast inward currents, corresponding to ectopic spikes, which occur in 4-AP. Carbenoxolone does not suppress large depolarizing IPSPs induced by tetanic stimulation. We conclude that in 4-AP, axonal gap junctions could, at least in principle, account in part for both the large IPSPs, and for the antidromic activity in pyramidal neurons.

High-frequency population oscillations are predicted to occur in hippocampal pyramidal neuronal networks interconnected by axoaxonal gap junctions

In hippocampal slices, high-frequency (125–333 Hz) synchronized oscillations have been shown to occur amongst populations of pyramidal neurons, in a manner that is independent of chemical synaptic transmission, but which is dependent upon gap junctions. At the intracellular level, high-frequency oscillations are associated with full-sized action potentials and with fast prepotentials. Using simulations of two pyramidal neurons, we previously argued that the submillisecond synchrony, and the rapid time-course of fast prepotentials, could be explained, in principle, if the requisite gap junctions were located between pyramidal cell axons. Here, we use network simulations (3072 pyramidal cells) to explore further the hypothesis that gap junctions occur between axons and could explain high-frequency oscillations. We show that, in randomly connected networks with an average of two gap junctions per cell, or less, synchronized network bursts can arise without chemical synapses, with frequencies in the experimentally observed range (spectral peaks 125–182 Hz). These bursts are associated with fast prepotentials (or partial spikes and spikelets) as observed in physiological recordings. The critical assumptions we must make for the oscillations to occur are: (i) there is a background of ectopic axonal spikes, which can occur at low frequency (one event per 25 s per axon); (ii) the gap junction resistance is small enough that a spike in one axon can induce a spike in the coupled axon at short latency (in the model, a resistance of 273 MΩ works, with an associated latency of 0.25 ms). We predict that axoaxonal gap junctions, in combination with recurrent excitatory synapses, can induce the occurrence of high-frequency population spikes superimposed on epileptiform field potentials.

Slow Oscillations (≤1 Hz) Mediated by GABAergic Interneuronal Networks in Rat Hippocampus

Perfusion of rat brain slices with low millimole CsCl elicits slow oscillations of </=1 Hz in hippocampal CA1 pyramidal neurons. These oscillations are GABAA receptor-mediated hyperpolarizations that permit a coherent fire-pause pattern in a population of CA1 neurons. They can persist without the activation of ionotropic glutamate receptors but require adenosine-dependent inhibition of glutamate transmission. In response to external Cs+, multiple interneurons in the CA1 region display rhythmic discharges that correlate with the slow oscillations in CA1 pyramidal neurons. The interneuronal discharges arise spontaneously from the resting potential, and their rhythmicity is regulated by periodic, GABAA receptor-mediated hyperpolarizations. In addition, interneurons show periodic partial spikes and neurobiotin coupling, and applications of known gap junctional uncouplers interrupt the Cs+-induced slow rhythm in both CA1 pyramidal neurons and interneurons. We propose that these slow oscillations originate from a GABAergic interneuronal network that interacts through reciprocal inhibition and possibly gap junctional connection.

‘NON‐SYNAPTIC’ MECHANISMS IN SEIZURES AND EPILEPTOGENESIS

The role of 'non-synaptic' mechanisms (i.e. those mechanisms that are independent of active chemical synpases) in the synchronization of neuronal activity during seizures and their possible contribution to chronic epileptogenesis are summarized. These 'non-synaptic' mechanisms include electrotonic coupling through gap junctions, electrical field effects (i.e. ephaptic transmission), and ionic interactions (e.g. increases in the extracellular concentration of K(+)). Several lines of evidence indicate that granule cells and pyramidal cells of the hippocampus, and probably other cortical neurons, can generate synchronized electrical activity after active chemical synaptic transmission has been blocked. This synchronized activity is sensitive to alterations in the size of the extracellular space, thus suggesting that electrical field effects and ionic mechanisms contribute to this synchronized activity. Recent studies also indicate that 'non-synaptic' synchronization is quite prominent early in development. Electrophysiological data from hippocampal and neocortical slices have led to a re-interpretation of the fast prepotentials (i.e. partial spikes) recorded in cortical pyramidal cells, suggesting that they may not be due to dendritic spike generation. Improvement in freeze-fracture ultrastructural techniques have led to a re-assessment of previous data on gap junctions in the nervous system and opened new approaches to the quantitative analysis and characterization of gap junctions on glia and neurons. Finally, new methods of dye/tracer coupling have the potential to provide a more rigorous basis for evaluating gap junctions and electrotonic communication between neurons in the mammalian central nervous system. Therefore, recent data continue to suggest that gap junctions and electrotonic coupling play an important role in neural integration, although additional studies using new techniques will be needed to address some of the controversial issues that have arisen over the last several decades.

Model of synchronized population bursts in electrically coupled interneurons containing active dendritic conductances.

We constructed a computer model of 128 interneurons, each with multiple dendritic branches and an axonal segment. The model neurons were interconnected by gap junctions between dendritic compartments, as are known to occur in rat and guinea-pig hilar interneurons. The model contained no excitatory synapses. In the presence of low-frequency spontaneous action potentials, the model generated synchronized population bursts, when gap junction resistance was 50 M omega and there were at least two gap junctions per neuron on average. Population bursts occurred only when the dendrites of model neurons were electrically excitable. Consistent with experiment, somatic hyperpolarization during the population burst uncovered partial spikes. In the model, partial spikes originated in electrically active dendrites driven by coupled dendrites. This model may account for population bursts in hilar interneurons that occur in 4-aminopyridine (4AP) together with blockers of GABAA and excitatory amino acid (EAA) receptors.

Multiple sites for the initiation of action potentials in neurons of the inferior mesenteric ganglion of the guinea-pig

Responses to nerve stimulation were recorded with intracellular microelectrodes from neurons in the inferior mesenteric ganglion of the guinea-pig. An event was recorded which was subthreshold for a somatic action potential but which had a short rise time and a rapid initial repolarization that gave the event a small but well-defined peak. This event was termed a “partial spike”. During repeated stimulation of the same nerve trunk, the fluctuations in the amplitude of the partial spike were small compared to those of evoked synaptic potentials. Stimulation of different nerve trunks evoked partial spikes of different amplitudes. When different nerve trunks were stimulated at short intervals between stimuli, one partial spike could occlude another partial spike. Antidromic responses could not be blocked by a preceding partial spike. This suggests that partial spikes are not initial segment spikes. Tubocurarine reversibly abolished partial spikes which is indicative of a synaptic origin. It is concluded that partial spikes result from action potentials initiated by synaptic potentials in the dendrites and which fail to generate somatic action potentials.

Ionic mechanisms underlying the firing properties of rat neonatal motoneurons studied in vitro

Ionic mechanisms underlying the firing properties of spinal motoneurons of neonatal rats (postnatal days 3–10) have been investigated using a hemisected, in vitro spinal cord preparation. These results demonstrate the presence of a high-threshold voltage-dependent calcium response and partial sodium-dependent spikes. The calcium current is evident during the falling phase of the action potential and is the major component of the after-depolarizing potential. The subsequent increase in intracellular calcium concentration activates a calcium-dependent potassium conductance (g K-ca), the major component of the after-hyperpolarizing potential. The g ca, by activating g K-ca, is the primary determinant of firing rate in neonatal motoneurons. For, when g ca, was blocked by Cd 2+, the interspike interval decreased, the maximum firing rate and the slope of the firing frequency-injected current relation increased. The calcium current is particularly robust during the first few postnatal days; during this period, tetrodotoxin resistant action potentials can be elicited by direct stimulation under control conditions. In animals older than 5 days such calcium spikes could be elicited only after decreasing g k with intracellular Cs + or extracellular tetraethylammonium. This was the case even when 1 mM of the bath CaCl 2 was replaced with BaCl 2. The rising phases of calcium spikes recorded from neurons in both age groups demonstrate several components suggesting the calcium spikes comprise several discrete events, which probably originate across the dendritic membrane. When g k was decreased by bath application of tetraethylammonium + and Cs +, neonatal motoneurons generated prolonged Ca-dependent spikes lasting for up to 6 s. Repolarization of Ca spikes occurred in two stages, the first was rapid (− 2.11 ±0.8 V/s, n = 6) but incomplete. The second, was slower (−0.01 ± 0.003 V/s, n = 5) and returned the membrane potential to the resting level after about 1–2 s. It is suggested that accumulation of extracellular potassium may contribute to the slow phase of repolarization. Motoneurons from the younger age group (3–5 days old) demonstrate all-or-none partial spikes rising from the after-depolarization of directly elicited sodium-dependent action potentials. Similar partial spikes were elicited from neurons from older animals during intracellular Cs + loading. The partial spikes had faster rates of rise than the tetrodotoxin-resistant spikes and were not seen after tetrodotoxin treatment, suggesting that they are sodium-dependent. These findings suggest that, in addition to the action potentials arising at the initial segment, neonatal motoneurons generate both sodium-dependent and calcium-dependent electroresponsive events at anatomically discrete sites, probably across the dendritic membrane. Further, in these immature cells an increase in intracellular calcium concentration activates a g k-ca which prolongs the after-hyperpolarization and thus gives the cells a low maximum firing rate compared to mature motoneurons, insuring a firing level well-matched to the properties of the immature sensory and motor systems.

Sodium-dependent regenerative responses in dendrites of axotomized motoneurons in the cat.

Ten days after extradural axotomy, partial spikes are found in greater than 20% of cat L7 motoneurons, while 15-21 days after axotomy the incidence increases to 60%. These responses are produced in excitable (hot) spots in the dendrites by synaptic excitation. Intracellular injection of QX-314, a lidocaine derivative and effective blocker of Na+ channels from within neurons, results in elimination of partial spikes before blocking somadendritic spikes. The action of QX-314 does not depend on changes in passive membrane properties or on changes in synaptic properties. Injections of Cs+ and Cl- ions rule out any major role for calcium, potassium, or chloride currents in the production of partial spikes. The partial spikes represent an unusual Na+-dependent dendritic phenomenon induced by axotomy when carried out relatively near the soma. It is reasonable to postulate that the partial spikes result from a higher concentration of Na+ channels in the dendrites. This may be the consequence of a high rate of production of Na+-channel proteins that are intended for the cut end of the axon; alternatively, they may result from the reflection from the cut end of such proteins produced at either a normal or an increased rate. These aberrant channels are inserted into somatic and dendritic membranes in higher concentrations than normal and, as well as producing local dendritic regions of low safety factor responsible for the partial spikes, also produce somadendritic spikes of unusually fast rise time and lower than usual threshold, which are relatively resistant to QX-314.

Ionic currents and firing patterns of mammalian vagal motoneurons In vitro

The electrophysiological properties of guinea-pig dorsal vagal motoneurons were studied in an in vitro slice preparation. Antidromic, orthodromic and direct stimulation of the neurons demonstrated that the action potential is comprised of several distinct components: a fast initial spike followed by afterdepolarization and an early and a late afterhyperpolarizations. The fast initial spike and the early afterhyperpolarization were blocked by tetrodotoxin and tetraethylammonium ions, respectively. The afterdepolarization (present on the falling phase of the spike) and the late afterhyperpolarization were blocked by the addition of ions known to block calcium conductance (CdCl 2, CoCl 2 or MnCl 2), indicating close association between these two potentials. Prolonged outward current injection through the recording electrode produced two different firing patterns, depending on the initial level of the membrane potential. From resting potential (usually − 60 mV) the firing pattern was characterized by a short train of action potentials appearing shortly after the onset of the depolarization step. By contrast, when the depolarization was delivered from a hyperpolarized membrane potential level, a short train of repetitive firing appeared after an initial delay of 300–400 ms. The membrane current responsible for this initial reduction in excitability was studied by means of a single-electrode voltage-clamp technique. The magnitude, direction and kinetics of such current flow are consistent with the presence of early potassium current ( I A ), partly inactive at the resting potential. Synaptic activation of vagal motoneurons could be obtained by electrical stimulation of the tissue surrounding the vagal nucleus or by direct activation of the vagal nerve. Perivagal stimulation generated excitatory and inhibitory synaptic potentials which could be reversed by shifting the membrane potential. Vagal nerve stimulation, in addition to the antidromic activation of the cells, generated depolarizing responses which were unitary in nature and did not show much sensitivity to shifts in membrane potential. Perivagal and vagal nerve-evoked depolarizations could generate action potentials as well as partial dendritic spikes. We conclude that spike electroresponsiveness in vagal motoneurons is generated by voltage-dependent Na + and Ca 2+ conductances. In addition, the Ca 2+-dependent current triggers a K + conductance which is responsible for modulating the firing frequency obtained from the normal resting level. A transient voltage-dependent K + current, similar to the A current, is responsible for the initial delay in the repetitive action potential trains elicited from hyperpolarized levels. The synaptic potentials evoked in these neurons resemble those observed in motoneurons following Ia stimulation and suggest that, at least in these motoneurons, excitatory postsynaptic-like potentials are generated, in part, by electroresponsive membrane properties of the postsynaptic membrane.

Synchronization without active chemical synapses during hippocampal afterdischarges.

Afterdischarges elicited by antidromic stimulation were studied in the CA1 region of transverse slices of rat hippocampus. The slices were incubated in a static bathing medium containing Mn2+ with lowered Ca2+ concentration, which blocked chemical synaptic transmission. Extracellular voltage recordings from micropipettes revealed afterdischarges of population spikes, which represent synchronized action potentials. Afterdischarge duration became longer with incubation time, occasionally lasting up to 9 s. Simultaneous recordings from three extracellular micropipettes indicated propagation of individual population spikes along the CA1 cell body layer during an afterdischarge. Intracellular recordings from pyramidal cells showed that action potentials were synchronized with population spikes. When initial action potentials of an afterdischarge were blocked by hyperpolarizing the impaled cell, subsequent action potentials remained synchronized with population spikes, indicating a synchronizing interaction between neurons. Intrasomatic neuronal recordings sometimes revealed partial action potentials that were also synchronized with population spikes. During afterdischarges, relatively weak (10 mV) hyperpolarization blocked both full and partial action potentials. These observations suggest that synchronized action potentials and partial spikes were initiated near the soma. Intracellular voltage recordings were referenced to a local extracellular microelectrode in order to determine more accurately the transmembrane potential in the presence of large population spikes .54). Such neuronal recordings revealed brief transmembrane depolarizations that occurred synchronously with population spikes. Large injected hyperpolarizing currents did not affect the brief transmembrane depolarizations. Similar differential voltage recordings across glial cell membranes did not show measurable brief transmembrane depolarizations during population spikes. However, slow glial depolarizations during afterdischarges indicated prolonged K+ accumulation and clearance in the extracellular space, which probably contributed to the hyperexcitability of pyramidal cells. During an afterdischarge, extracellular potentials were recorded simultaneously from six micropipettes arranged perpendicularly to the stratum pyramidale. Current source-density analysis indicated a sink in the cell body layer during population spikes and a corresponding source in distal dendrites. It is concluded that an electrical field effect depolarizes and thus synchronizes neurons during these afterdischarges.(ABSTRACT TRUNCATED AT 400 WORDS)

Electrophysiological abnormalities in chronic epileptogenic foci: an intracellular study

Intracellular recordings were performed from chronic cobalt epileptogenic foci in rats. Two main features were found: (a) in 29% of the neurons, large depolarizations of long duration occurred coinciding with cortical surface paroxysms; and (b) in 13% of the neurons, partial spikes, whose amplitude varied little with membrane DC potential changes, occurred. It is suggested that these phenomena have a dendritic origin, and possible cellular mechanisms underlying the generation of paroxysmal events are discussed.

Action potential generation by turtle cortical neurons. Dendritic and somatic spikes

Action potentials of neurons of the turtle general cortex and the pattern of their generation were studied by an intracellular recording method. Besides the complete action potential, the cells also generate partial spikes of varied amplitude which compose the complete action potential. The threshold of generation and the discrete amplitude of each partial spike are not strictly constant but they fluctuate gradually and spontaneously within certain limits without any change in membrane potential of the cell. Somatic and dendritic spikes are distinguished. The trigger zones of the latter are located at various distances from the soma. During orthodromic activation of cortical neurons dendritic spikes are generated consecutively and spread to the some electrotonically with a decrement. They are the immediate cause of generation of the somatic spike.

Crossed effects on central vestibular neurons in the horizontal canal system of the frog.

1. Neuronal discharges were recorded extracellularly in curarized frogs (Rana esculenta) with glassmicropipettes. Vestibular neurons, which were activated by ipsilateral horizontal angular acceleration and suppressed during deceleration (type 1) were found mainly in the medial part of the ventral vestibular nucleus. 2. In contrast to type 1 vestibular neuron of the cat, this type of neuron in the frog was not suppressed, but facilitated by electrical stimulation of the contralateral vestibular nerve, irrespective of stimulus intensity. In most cases the facilitation was subthreshold for producing full action potentials when single stimulation was applied. 3. Single shocks applied to the contralateral vestibular nerve evoked small negative field potentials (maximum peak amplitude, 300 μV) in the vestibular nucleus. The threshold for the potential was 2.5–3.0 times threshold for the field potential recorded in the vestibular nucleus on the stimulated side. The latency ranged from 5.0 to 8.0 msec. 4. Intracellular recordings were obtained from the neurons in the medial part of the ventral vestibular nucleus. These neurons exhibited monosynaptic EPSP in response to ipsilateral vestibular nerve stimulation. Following electrical stimulation of the contralateral vestibular nerve polysynaptic EPSPs were found with latencies between 5.0 and 10.0 msec and no IPSPs were detected. 5. The EPSPs induced by single supramaximal shocks to the contralateral vestibular nerve were in most cases not large enough to evoke full action potentials. In more than half of the cells recorded from, partial spikes were superimposed on the EPSPs. 6. In the vestibular neurons which send axons to the peripheral receptor organs, EPSPs were induced by stimulation of the contralateral vestibular nerve. There were no highly significant differences between the time courses of these EPSPs and those described in 4.

Electrophysiology of “epileptic” neurons: Spike generation

1. 1. Intracellular recordings were obtained from 74 neurons in penicillin-induced epileptogenic foci on cat pericruciate cortex in order to study the mechanisms of spike generation in “epileptic” neurons. Spontaneous and evoked spike discharge and spiking associated with the large depolarization shifts (DSs) which occur in cells during inter-ictal epileptiform surface events were examined. 2. 2. Orthodromic and antidromic spike generation (14 PT cells) was generally normal in that the usual IS-SD spike components were demonstrated and spikes arose from stable firing levels. Direct depolarizing current pulses produced high frequency spiking without DS generation. 3. 3. Antidromic spikes elicited during a DS did not alter the DS amplitude or time course. Antidromic invasion failed during the peak of the DS in some cells, due to cathodal blockade and also during periods of high frequency DS spike generation, presumably because of refractoriness or collision in the axon. 4. 4. Peculiarities of spike generation occurred as follows: (1) In 19 cells the first DS spike sometimes arose from a membrane potential 2–10 mV below the usual firing level without a preceding slow pre-potential. (b) In 4 of these cells orthodromic spikes not associated with the DS also occasionally arose from below threshold and appeared to have more than one mode of onset. (c) In 3 neurons, intracellular negative or negative-positive spikes occurred during surface epileptiform discharges. (d) Partial spikes or “spiklets” in addition to spikes of normal amplitude were noted in 18 cells of the focus and 4 of 28 control cells from normal cortex. 5. 5. These pecularities of spike generation suggest that in proportion of neurons of the penicillin epileptogenic focus, impulses originate from trigger zones distant from the soma, presumably from dendrites. These data raise the possibility that dendritic spike generation may be a general property of “epileptic” neurons.

Activation of "secondary" impulse trigger sites in hippocampal neurones.

INTRACELLULARLY recorded spike potentials of hippocampal neurones occasionally exhibit fast prepotentials superimposed on spontaneous slow depolarizations and excitatory postsynaptic potentials evoked by stimulation of the subiculum1. Although the origin and nature of fast prepotentials have not been determined, available data suggest that they represent electrotonic reflexions of partial spikes evoked in dendrites by pathways arising in the subiculum1. Two investigations which are summarized here have permitted further analysis of fast prepotentials as well as the definition of some of the factors conditioning their development in hippocampal neurones. The results are illustrated in contrast with findings in elements which did not exhibit fast prepotentials during a long period of spontaneous activity or in response to low frequency (1/sec) stimulation of the perforant pathway in the subiculum.