Real-time closed-loop electrophysiology: towards new frontiers in in vitro investigations in the neurosciences.

Reflected at any level of organization of the central nervous system, most of the processes ranging from ion channels to neuronal networks occur in a closed loop, where the input to the system depends on its output. In contrast, most in vitro preparations and experimental protocols operate autonomously, and do not depend on the output of the studied system. Thanks to the progress in digital signal processing and real-time computing, it is now possible to artificially close the loop and investigate biophysical processes and mechanisms under increased realism. In this contribution, we review some of the most relevant examples of a new trend in in vitro electrophysiology, ranging from the use of dynamic-clamp to multi-electrode distributed feedback stimulation. We are convinced these represents the beginning of new frontiers for the in vitro investigation of the brain, promising to open the still existing borders between theoretical and experimental approaches while taking advantage of cutting edge technologies.

Transient rhythmic network activity in the somatosensory cortex evoked by distributed input in vitro.

The initiation and maintenance of physiological and pathophysiological oscillatory activity depends on the synaptic interactions within neuronal networks. We studied the mechanisms underlying evoked transient network oscillation in acute slices of the adolescent rat somatosensory cortex and modeled its underpinning mechanisms. Oscillations were evoked by brief spatially distributed noisy extracellular stimulation, delivered via bipolar electrodes. Evoked transient network oscillation was detected with multi-neuron patch-clamp recordings under different pharmacological conditions. The observed oscillations are in the frequency range of 2-5 Hz and consist of 4-12 mV large, 40-150 ms wide compound synaptic events with rare overlying action potentials. This evoked transient network oscillation is only weakly expressed in the somatosensory cortex and requires increased [K+]o of 6.25 mM and decreased [Ca2+]o of 1.5 mM and [Mg2+]o of 0.5 mM. A peak in the cross-correlation among membrane potential in layers II/III, IV and V neurons reflects the underlying network-driven basis of the evoked transient network oscillation. The initiation of the evoked transient network oscillation is accompanied by an increased [K+]o and can be prevented by the K+ channel blocker quinidine. In addition, a shift of the chloride reversal potential takes place during stimulation, resulting in a depolarizing type A GABA (GABAA) receptor response. Blockade of alpha-amino-3-hydroxy-5-methyl-4-isoxazole-proprionate (AMPA), N-methyl-D-aspartate (NMDA), or GABA(A) receptors as well as gap junctions prevents evoked transient network oscillation while a reduction of AMPA or GABA(A) receptor desensitization increases its duration and amplitude. The apparent reversal potential of -27 mV of the evoked transient network oscillation, its pharmacological profile, as well as the modeling results suggest a mixed contribution of glutamatergic, excitatory GABAergic, and gap junctional conductances in initiation and maintenance of this oscillatory activity. With these properties, evoked transient network oscillation resembles epileptic afterdischarges more than any other form of physiological or pathophysiological neocortical oscillatory activity.

A modified roller tube technique for organotypic cocultures of embryonic rat spinal cord, sensory ganglia and skeletal muscle.

The roller tube technique as initially described in the literature in 1981, was modified in several aspects for the coexplantation of embryonic rat spinal cord with attached dorsal root ganglia and skeletal muscle from newborn rats. The high metabolic activity of this coculture system required a particular culturing protocol to stabilize pH and osmotic pressure. The appropriate adjustment of the partial pressure of carbon dioxide gas in the incubator proved to be essential for the control of the pH within narrow limits (7.3 +/- 0.1). The adjustment of the osmotic pressure of the medium (290-300 mOsm) improved the growth of the cultures considerably. Roller drum speed was set to 120 revolutions per hour for enhanced flattening of the culture. A simple rating system was used to evaluate neuronal and non-neuronal outgrowth under different modifications of the culture system. Furthermore, morphological and electrophysiological criteria were defined for evaluating individual neurons. The technique described insures the growth of long-term organotypic cocultures of spinal cord, sensory ganglia and skeletal muscle.

Wiring diagrams of functional connectivity in monosynaptic reflex arcs of the spinal cord.

The direct functional connections between Ia and group II spindle afferent fibers from the cat medial gastrocnemius muscle and their homonymous motoneurons were examined in 10 acute experiments. Trains of stretch-evoked impulses from as many as 20 undivided sensory fibers were recorded simultaneously from 5 dorsal root filaments, as well as the corresponding excitatory postsynaptic potentials (EPSPs) they elicited in 10-20 motoneurons. Spike-triggered averaging [13] of these tape-recorded signals revealed the functional connections (or non-connections) between each Ia or group II afferent fiber and each motoneuron. Wiring diagrams constructed from these data indicate that the probability of a functional connection between an afferent fiber and a motoneuron decreases with the size of either and with the distance between the entry point of the afferent fiber and the motoneuron.

Simulation of action potential propagation in terminal arborizations of cultured sensory ganglia.

In order to study possible integrative properties of axon terminations, impulse propagation in arbitrarily complex terminal arborizations is simulated using a general purpose network analysis program. Detailed, anatomically based models are constructed from HRP-filled arborizations of sensory ganglia impinging upon motoneurons grown in an organotypic rat spinal cord culture. On assuming an excitable membrane the action potential propagates into all the ramifications. If some branches are assumed to be inexcitable, the electrotonically propagated potential may not depolarize the synaptic endings sufficiently to release transmitter. Under such conditions only part of the morphologically found synapses are expected to release transmitter following stimulation of sensory neurons.

Miniature excitatory synaptic currents corrected for dendritic cable properties reveal quantal size and variance.

1. Non-NMDA (N-Methyl-D-aspartate) receptor mediated miniature excitatory synaptic currents (mEPSCs) were recorded from motoneurons in organotypic cultures of embryonic rat spinal cord. 2. Amplitude histograms of mEPSCs were unimodal and skewed toward larger events. The mean of the modes of the amplitude histograms was -18 pA with a maximal amplitude range of -4 to -160 pA for individual mEPSCs. 3. Current transients to a short voltage pulse were used to estimate the passive cable parameters of the motoneurons. The mean membrane time constant (tau) and the mean electrotonic length (L) were 20 and 0.96 ms, respectively. 4. The amplitudes of the mEPSCs were corrected for imperfect space and voltage clamp. The resulting amplitude histograms could be fitted by the sum of two Gaussian curves, revealing a mean quantal size of -48 pA with a coefficient of variation (cv) of 0.28. 5. Our data suggest that quantal size and its variance are masked by the cable properties of the neurons and that simultaneous release of elementary quanta occurs occasionally.

Modeling action potential initiation and back-propagation in dendrites of cultured rat motoneurons.

Regardless of the site of current injection, action potentials usually originate at or near the soma and propagate decrementally back into the dendrites. This phenomenon has been observed in neocortical pyramidal cells as well as in cultured motoneurons. Here we show that action potentials in motoneurons can be initiated in the dendrite as well, resulting in a biphasic dendritic action potential. We present a model of spinal motoneurons that is consistent with observed physiological properties of spike initiation in the initial segment/axon hillock region and action potential back-propagation into the dendritic tree. It accurately reproduces the results presented by Larkum et al. on motoneurons in organotypic rat spinal cord slice cultures. A high Na+-channel density of Na = 700 mS/cm2 at the axon hillock/initial segment region was required to secure antidromic invasion of the somato-dendritic membrane, whereas for the orthodromic direction, a Na+-channel density of Na = 1,200 mS/cm2 was required. A "weakly" excitable (Na = 3 mS/cm2) dendritic membrane most accurately describes the experimentally observed attenuation of the back-propagated action potential. Careful analysis of the threshold conditions for action potential initiation at the initial segment or the dendrites revealed that, despite the lower voltage threshold for spike initiation in the initial segment, an action potential can be initiated in the dendrite before the initial segment fires a spike. Spike initiation in the dendrite depends on the passive cable properties of the dendritic membrane, its Na+-channel density, and local structural properties, mainly the diameter of the dendrites. Action potentials are initiated more easily in distal than in proximal dendrites. Whether or not such a dendritic action potential invades the soma with a subsequent initiation of a second action potential in the initial segment depends on the actual current source-load relation between the action potential approaching the soma and the electrical load of the soma together with the attached dendrites.

Computation of action potential propagation and presynaptic bouton activation in terminal arborizations of different geometries.

Action potential propagation in axons with bifurcations involving short collaterals with synaptic boutons has been simulated using SPICE, a general purpose electrical circuit simulation program. The large electrical load of the boutons may lead to propagation failure at otherwise uncritical geometric ratios. Because the action potential gradually fails while approaching the branch point, the electrotonic spread of the failing action potential cannot depolarize the terminal boutons above an assumed threshold of 20 mV (Vrest = 0 mV) for the presynaptic calcium inflow, and therefore fails to evoke transmitter release even for boutons attached at short collaterals. For even shorter collaterals the terminal boutons can again be activated by the spread of passive current reflected at the sealed end of the bouton which increases the membrane potential above firing threshold. The action potential is then propagated in anterograde fashion into the main axon and may activate the terminal bouton on the other collateral. Differential activation of the synaptic boutons can be observed without repetitive activation of the main axon and with the assumption of uniform membrane properties. Axon enlargements above a critical size at branch points can increase the safety factor for propagation significantly and may serve a double function: they can act both as presynaptic boutons and as boosters, facilitating invasion of the action potential into the terminal arborizations. The architecture of the terminal arborizations has a profound effect on the activation pattern of synapses, suggesting that terminal arborizations not only distribute neural information to postsynaptic cells but may also be able to process neural information presynaptically.

Simulation of action potential propagation in complex terminal arborizations.

Action potential propagation in complex terminal arborizations was simulated using SPICE, a general purpose circuit simulation program. The Hodgkin-Huxley equations were used to simulate excitable membrane compartments. Conduction failure was common at branch points and regularly spaced boutons en passant. More complex arborizations had proportionally more inactive synapses than less complex arborizations. At lower temperature the safety factor for impulse propagation increased, reducing the number of silent synapses in a particular arborization. Small structural differences as well as minute changes in the discharge frequency of the action potential resulted in very different activation patterns of the arborization and terminal boutons. The results suggest that the structural diversity of terminal arborizations allows a wide range of presynaptic information processing. The results from this simulation study are discussed in the context of experimental results on the modulation of synaptic transmission.

A comparison of homonymous and heteronymous connectivity in the spinal monosynaptic reflex arc of the cat.

Multi-unit spike triggered averaging was used to determine functional connectivity between spindle afferent fibers from the medial gastrocnemius muscle and the motoneurons innervating the medial (homonymous connections) and the lateral gastrocnemius-soleus muscle (heteronymous connections). As many as 288 possible connections between 24 motoneurons and 12 afferent fibers were studied in single, acute experiments. The influences of morphological and topographical factors, as well as of motoneuron species on functional connectivity were analysed. The probability that a motoneuron would receive functional connections from a given population of afferent fibers was related to its size and its proximity to the spinal entry level of the afferent fibers. The faster the axonal conduction velocity of the motoneuron (i.e. the larger the motoneuron) and the closer its location to the entry zone of the afferent fibers, the higher was its probability of receiving functional connections. The greater the conduction velocity (i.e. diameter) of a stretch receptor afferent fiber, the higher was its probability of making functional connections with motoneurons. These relationships were qualitatively similar for homonymous and heteronymous connections. 58% (233/399) of the Ia and group II afferents (combined) had functional connections with homonymous motoneurons, 32% (75/234) with heteronymous motoneurons. However, homonymous and heteronymous motoneurons of similar sizes were equally likely to receive functional connections when located at the same craniocaudal level. Differences in the locations and mean sizes of homonymous and heteronymous motoneurons however, cannot account completely for the observed overall differences in homonymous and heteronymous connectivity.

Miniature excitatory postsynaptic potentials in embryonic motoneurons grown in slice cultures of spinal cord, dorsal root ganglia and skeletal muscle.

Miniature excitatory postsynaptic potentials (mEPSPs) were recorded in motoneurons grown in organotypic cocultures of embryonic rat spinal cord, dorsal root ganglia and muscle in the presence of TTX. The motoneurons were electrically compact with a mean electrotonic length of 0.6. Spontaneous EPSPs were found in most of these motoneurons. With TTX the large EPSPs disappeared, whereas in more than half of the experiments mEPSPs persisted with a range in size of 1 to 4 mV (mean: 2.1 mV), probably originating from the spontaneous release of single vesicles. The net inward charge transfer at the soma ranged from 0.12 to 0.34 pC. The mEPSPs were heterogeneous in size even within pools of potentials that were homogeneous in shape. They had similar shapes and amplitudes as the smallest spontaneous unitary EPSPs mediated by presynaptic impulses, suggesting that for the smallest afferents not more than one vesicle was released per afferent impulse. Both the miniature and the TTX-sensitive EPSPs were readily blocked by the glutamate antagonist DNQX.

Analysis of synaptic transmission at single identified boutons on rat spinal neurons in culture.

The spatial organization of receptor channels has a major influence on the speed and possible plasticity of synaptic signal transmission. We have studies glutamatergic synapses on neurons in organotypic cultures of rat spinal cord. In order to avoid the problems related to the analysis of currents of unknown origin within a neuron, we chose to examine the functional properties of single identified synapses. Iontophoretic mapping of the cell surface revealed hot spots of high glutamate sensitivity coincident with presynaptic boutons stained with the dye FM 1-43. Local application of KCl to these sites caused bursts of synaptic release. Hot spots typically consisted of 330 receptors with an average single-channel conductance of 8.3 pS. Evoked synaptic currents involved only about 40-50 receptors and nevertheless showed characteristics of saturation. This suggests that glutamate receptor clusters at sites of presynaptic terminals are organized into well separated subclusters opposite release sites.

Practical guidance for testing the accuracy of deconvolution results from quantal analysis.

A Monte Carlo study was carried out to test the reliability of the Maximum Likelihood Estimator (MLE) approach for quantal analysis. This widely used statistical method was applied to extract a finite mixture of Gaussian distributions from simulated data. The data were generated by convolving a distribution of discrete amplitude steps (multiples of a unitary step Q) with Gaussian noise of various standard deviations (sigma n). Our results offer practical guidance on when to use the MLE, taking into account the determining parameters: signal to noise ratio (Q/sigma n, the most important parameter), number of samples collected and the number of components (k). For a given set of parameters the algorithm always converged to the "true" values, never converged to the "true" values or converged in only a fraction of cases to the "true" values. The behavior of the fitting routine in the parameter space is displayed in contour plots. These contour plots can be used as a guide to test the accuracy of deconvolution results.

Action potential propagation through embryonic dorsal root ganglion cells in culture. I. Influence of the cell morphology on propagation properties.

1. In this and the companion paper the reliability of action potential (AP) propagation through dorsal root ganglion (DRG) cells was investigated. Experimental data were collected from DRG cells of embryonic rat slice cultures of the spinal cord. A field stimulation electrode was used to elicit an AP in the axon. The propagated AP or, in case of conduction block, its electronic residue (ER), was measured intracellularly in the soma of the DRG cell. 2. The morphological and electrophysiological data combined with published data from voltage-clamp studies were taken to implement a compartmental computer model, which allows a precise description of the propagating AP and the channel kinetics at any point along the axon. 3. The safety factor for conduction was found to be low. Thus failures of AP invasion of the DRG cell soma could occur at sites of impedance mismatch when a hyperpolarizing current was applied, a second stimulus felt into the relative refractory period of the first, or when the axon was repetitively stimulated. 4. The ERs of the failed APs had discrete amplitude levels, suggesting that the failures were always caused at the same site along the axon. These sites of low safety factor were found to be the branch point in the unipolar DRG cell and the entrance of the stem piece into the soma in both cell types, the bipolar as well as the unipolar. 5. A systematic comparison of bipolar and unipolar DRG cells showed that the AP conduction through the latter is more reliable. For large cell bodies, the unipolar configuration is needed for save conduction. 6. Conduction through unipolar DRG cells is faster than through bipolar cells because the electrical load of the soma is masked by the high-resistive stem piece. The length of this stem piece is correlated inversely to the delay caused at the branch point, as the electrical load of the soma is more efficiently masked by a long stem piece.

Action potential propagation through embryonic dorsal root ganglion cells in culture. II. Decrease of conduction reliability during repetitive stimulation.

1. The reliability of the propagation of action potentials (AP) through dorsal root ganglion (DRG) cells in embryonic slice cultures was investigated during repetitive stimulation at 1-20 Hz. Membrane potentials of DRG cells were recorded intracellularly while the axons were stimulated by an extracellular electrode. 2. In analogy to the double-pulse experiments reported previously, either one or two types of propagation failures were recorded during repetitive stimulation, depending on the cell morphology. In contrast to the double-pulse experiments, the failures appeared at longer interpulse intervals and usually only after several tens of stimuli with reliable propagation. 3. In the period with reliable propagation before the failures, a decrease in the conduction velocity and in the amplitude of the afterhyperpolarization (AHP), an increase in the total membrane conductance, and the disappearance of the action potential "shoulder" were observed. 4. The reliability of conduction during repetitive stimulation was improved by lowering the extracellular calcium concentration or by replacing the extracellular calcium by strontium. The reliability of conduction decreased by the application of cadmium, a calcium channel blocker, 4-amino pyridine, a fast potassium channel blocker, or apamin or muscarine, the blockers of calcium-dependent potassium channels. The reliability of conduction was not effected by blocking the sodium potassium pump with ouabain or by replacing extracellular sodium with lithium. 5. In the period with reliable propagation cadmium, apamin, and muscarine reduced the amplitude of the AHP. The shoulder of the action potential was more pronounced and not sensitive to repetitive stimulation when extracellular calcium was replaced by strontium. It disappeared when cadmium was applied. 6. In DRG somata changes of the intracellular Ca2+ concentration were monitored by measuring the fluorescence of the Ca2+ indicator Fluo-3 with a laser-scanning confocal microscope. During repetitive stimulation, an accumulation of intracellular calcium occurred that recovered very slowly (tens of seconds) after the AP trains. 7. Computer model simulations performed in analogy to the experimental protocols produced conduction failures during repetitive stimulation only when the calcium currents during the APs were reduced. 8. From these findings it is concluded that conduction failures during repetitive stimulation are dependent on an accumulation of intracellular calcium leading to an inactivation of calcium currents, combined with small contributions of an accumulation of extracellular potassium and a summation of slow potassium conductances.

Electronic structure of motoneurons in spinal cord slice cultures: a comparison of compartmental and equivalent cylinder models.

1. Voltage-clamp, current-clamp, and morphological data were obtained from visually identified motoneurons in organotypic cocultures of rat embryonic spinal cord, dorsal root ganglia, and skeletal muscle. The cells were injected with Biocytin during whole-cell patch-clamp recordings and stained with horseradish peroxidase. 2. The somata and dendritic trees of the cells were reconstructed with a semiautomatic reconstruction system. The motoneurons had a common multipolar shape. An elliptic soma gave rise to 3-9 stem dendrites with a mean diameter of 2.5 +/- 0.9 (SD) micron terminating in 24 +/- 7 dendritic endings. The mean total dendritic path length was 3,306 +/- 1,075 microns. The mean total membrane surface area was 15,594 +/- 10,404 microns 2 with a dendritic to somatic membrane surface area ratio of 3.4 +/- 1.4 (n = 7 cells). 3. The ratio between the sum of the diameters of the two daughter branches and the diameter of the parental branch each raised to the 3/2 power at all branch points was 1.3 +/- 0.28 (n = 8 cells). The dendritic trees of the cells tapered continuously from the soma to the distal ends. The mean normalized dendritic trunk parameter of all cells was 0.62 +/- 0.22. 4. The motoneurons had a mean input resistance RN of 498 +/- 374 M delta, a mean membrane time constant (tau m) of 22 +/- 4.6 ms, and a mean dendritic dominance (rho) of 2.7 +/- 0.86 (n = 5 cells). The mean electronic length (L) calculated from tau m and the slowest voltage-clamp time constant (tau VC1) was 0.7 +/- 0.04 (n = 7 cells). 5. The specific membrane capacitance (Cm) estimated from the charge of the capacitive current during a voltage step and the total membrane surface area was 1.08 +/- 0.3 microF/cm2 (n = 6 cells). 6. Compartmental computer models were constructed of individual cells. Experimental and simulated voltage transients were matched with Cm = 1 microF/cm2, a uniform membrane resistivity (Rm) = tau m/Cm and a cytosolic resistivity (Ri) of 308 +/- 39 omega.cm (n = 3 cells). 7. The mean electrotonic length of the dendritic paths was 0.83 +/- 0.2 (n = 5 cells). The mean input resistance at the dendritic terminals (RT) was 1,413 +/- 260 M omega. Synaptic conductances were applied at all distal dendritic compartments of the model cells. The resulting synaptic currents were calculated at the input site and at the soma. The mean transient current attenuation ratio was 4.7 +/- 1.7 under idealized voltage-clamp conditions.(ABSTRACT TRUNCATED AT 400 WORDS)

Synaptic plasticity in dissociated hippocampal cultures: pre- and postsynaptic contributions.

The distinction between pre- or postsynaptic expression of synaptic plasticity is difficult to make, unless the postsynaptic receptors can be investigated in isolation. We have studied single synaptic contacts in dissociated cultures of rat hippocampus. The reaction of postsynaptic receptor assemblies to the induction of synaptic plasticity was measured and compared with changes in the rate of spontaneous miniature excitatory postsynaptic currents (mEPSCs), which can reflect changes in the transmitter release mechanism. The response of a receptor assembly to locally applied exogenous glutamate was measured before and after synchronized application of glutamate and a train of postsynaptic depolarizations ('pairing'). Pairing induced a variety of changes: (i) the majority of the receptor assemblies showed no change in their response to glutamate before and after pairing; (ii) the postsynaptic current due to exogenous glutamate showed a rapid increase in five out of 26 cases. This was not due to changes in the single channel conductance; (iii) the rate of mEPSCs increased, if it had previously been below 25 Hz; (iv) the rate of mEPSCs decreased, if it had previously been above 25 Hz. Effects 2 and 3 were blocked by antagonists of NMDA receptors. These findings provide direct evidence for an increase of the number of glutamate receptors at a subset of the investigated postsynaptic sites during synaptic potentiation.