Recurrent Synapses Research Articles

Mechanisms of memory-supporting neuronal dynamics in hippocampal area CA3.

Hippocampal CA3 is central to memory formation and retrieval. Although various network mechanisms have been proposed, direct evidence is lacking. Using intracellular Vm recordings and optogenetic manipulations in behaving mice, we found that CA3 place-field activity is produced by a symmetric form of behavioral timescale synaptic plasticity (BTSP) at recurrent synapses among CA3 pyramidal neurons but not at synapses from the dentate gyrus (DG). Additional manipulations revealed that excitatory input from the entorhinal cortex (EC) but not the DG was required to update place cell activity based on the animal's movement. These data were captured by a computational model that used BTSP and an external updating input to produce attractor dynamics under online learning conditions. Theoretical analyses further highlight the superior memory storage capacity of such networks, especially when dealing with correlated input patterns. This evidence elucidates the cellular and circuit mechanisms of learning and memory formation in the hippocampus.

The Interplay of Synaptic Plasticity and Scaling Enables Self-Organized Formation and Allocation of Multiple Memory Representations.

It is commonly assumed that memories about experienced stimuli are represented by groups of highly interconnected neurons called cell assemblies. This requires allocating and storing information in the neural circuitry, which happens through synaptic weight adaptations at different types of synapses. In general, memory allocation is associated with synaptic changes at feed-forward synapses while memory storage is linked with adaptation of recurrent connections. It remains, however, largely unknown how memory allocation and storage can be achieved and the adaption of the different synapses involved be coordinated to allow for a faithful representation of multiple memories without disruptive interference between them. In this theoretical study, by using network simulations and phase space analyses, we show that the interplay between long-term synaptic plasticity and homeostatic synaptic scaling organizes simultaneously the adaptations of feed-forward and recurrent synapses such that a new stimulus forms a new memory and where different stimuli are assigned to distinct cell assemblies. The resulting dynamics can reproduce experimental in-vivo data, focusing on how diverse factors, such as neuronal excitability and network connectivity, influence memory formation. Thus, the here presented model suggests that a few fundamental synaptic mechanisms may suffice to implement memory allocation and storage in neural circuitry.

Multiassociative Memory: Recurrent Synapses Increase Storage Capacity

The connection density of nearby neurons in the cortex has been observed to be around 0.1, whereas the longer-range connections are present with much sparser density (Kalisman, Silberberg, & Markram, 2005 ). We propose a memory association model that qualitatively explains these empirical observations. The model we consider is a multiassociative, sparse, Willshaw-like model consisting of binary threshold neurons and binary synapses. It uses recurrent synapses for iterative retrieval of stored memories. We quantify the usefulness of recurrent synapses by simulating the model for small network sizes and by doing a precise mathematical analysis for large network sizes. Given the network parameters, we can determine the precise values of recurrent and afferent synapse densities that optimize the storage capacity of the network. If the network size is like that of a cortical column, then the predicted optimal recurrent density lies in a range that is compatible with biological measurements. Furthermore, we show that our model is able to surpass the standard Willshaw model in the multiassociative case if the information capacity is normalized per strong synapse or per bits required to store the model, as considered in Knoblauch, Palm, and Sommer ( 2010 ).

A Bio-Electrical Tornado in The Hippocampus: Mechanisms of Temporal Lobe Epilepsy

This review summarizes the structural and electrophysiological changes in the epileptic hippocampus in various animal models of epilepsy. Tornado hypothesis of epileptic hippocampus, states that there are many progressive detrimental changes in the hippocampus of the epileptic rats. PTZ, PTZ kindling, pilocarpine, electrical kindling, kainic acid, hyperthermia and in vitro models are taken as primary animal epilepsy models in this review. Paired pulse inhibition and GABAergic transmission is decreased in dentate gyrus (DG), CA3 and CA1 regions of the epileptic hippocampus. There is interneuron loss in the hilus, DG and CA3, while excitatory input from perforant path (PP) is increased and this is fed into the CA3 area by mossy fibers (MF) which have excessive sprouting that forms novel recurrent synapses with the pyramidal cells of CA3. Gate function of DG is either impaired or lost. CA3 area of hippocampus becomes an epileptic focus and sends ictal discharges. These ictal discharges are carried into CA1 by Schaffer collaterals and then into entorhinal cortex (EC). The input from EC is amplified in the hippocampal circuitry which is fed into hippocampus from EC into DG again successively. Thus a weak electrical input into the hippocampal formation results in an amplified signal back into EC. Since the basic ultrastructural and electrophysiological feed-back control mechanisms are impaired, this electrical tornado cannot be compensated for and an epileptic amplified ictal discharge spreads to the limbic system and other adjacent structures of the brain. Eventually the hippocampal circuitry, that has developed a vicious circle, becomes a bio-electrical amplifier which triggers an electrical tornado, under certain bio-chemical conditions.

Astrocytes Modulate Local Field Potential Rhythm?

Astrocytes Modulate Local Field Potential Rhythm?

Anticipative tracking in two-dimensional continuous attractor neural networks

Time delays exist pervasively in neural information processing. The brain needs to compensate for these delays in order to extract motion information in time [1]. Experimental data reveal that in tracking moving stimuli, neural systems generate anticipative responses to compensate for delays [2], but how does the brain achieves this goal remains poorly understood. Continuous attractor neural networks (CANNs) are widely used as a canonic model to describe the encoding of continuous features, such as head-direction, moving direction, orientation or spatial location of an object, in the brain. A CANN holds a continuous family of localized stationary states, called bumps. These bumps form a sub-manifold in the network state space, on which the system is neutrally stable, and this neutral stability endows the network with the capacity of tracking moving inputs smoothly [3]. Although a CANN is able to track a moving input, its reaction time is always delayed with respect to the instant position of the input, due to that neurons responding to external input and neuronal interaction via recurrent synapses consume time. In this study, by incorporating spike frequency adaptation (SFA), a negative feedback mechanism, in the network dynamics, we find that a CANN is able to track a moving input anticipatively. We consider a two-dimensional CANN, mimicking the neural circuit of place cells [4]. With theoretical analysis and numerical simulation, we systematically explore the dynamics of a 2D CANN, and find that: 1) in the absence of SFA or in the presence of weak SFA, the CANN holds static bump states; 2) when SFA is sufficiently strong, the CANN generates a traveling wave, in which a bump moves spontaneously in the network without relying on external drive. The speed of the traveling wave is fully determined by the network parameters, which we call the intrinsic speed vint of the network; 3) in response to a moving input with speed vext, the interplay between the intrinsic mobility of the network and the speed of the external drive determines the network tracking performance, that is, a) when vext> vint, the network state lags behind the external input; and b) when vext< vint, the network state leads the external input. Interestingly, we find that the anticipative time of the network, given by s/vext with s the leading distance, is approximately a constant for a wide range of input speed. This property justifies the experimental finding that in a given brain area, the anticipative time is approximately a constant independent of input speed. By constructing a hierarchical model with multiple CANNs, our model also reproduces the experimental finding that along the signal pathway, the anticipative time of neurons increases. Our study reveals that a neural system can utilize the intrinsic dynamics of a neural circuit to implement anticipative responses to moving inputs. It sheds light on our understanding of how the brain processes motional information in a timely manner.

Intrinsic mechanisms stabilize encoding and retrieval circuits differentially in a hippocampal network model.

Acetylcholine regulates memory encoding and retrieval by inducing the hippocampus to switch between pattern separation and pattern completion modes. However, both processes can introduce significant variations in the level of network activity and potentially cause a seizure-like spread of excitation. Thus, mechanisms that keep network excitation within certain bounds are necessary to prevent such instability. We developed a biologically realistic computational model of the hippocampus to investigate potential intrinsic mechanisms that might stabilize the network dynamics during encoding and retrieval. The model was developed by matching experimental data, including neuronal behavior, synaptic current dynamics, network spatial connectivity patterns, and short-term synaptic plasticity. Furthermore, it was constrained to perform pattern completion and separation under the effects of acetylcholine. The model was then used to investigate the role of short-term synaptic depression at the recurrent synapses in CA3, and inhibition by basket cell (BC) interneurons and oriens lacunosum-moleculare (OLM) interneurons in stabilizing these processes. Results showed that when CA3 was considered in isolation, inhibition solely by BCs was not sufficient to control instability. However, both inhibition by OLM cells and short-term depression at the recurrent CA3 connections stabilized the network activity. In the larger network including the dentate gyrus, the model suggested that OLM inhibition could control the network during high cholinergic levels while depressing synapses at the recurrent CA3 connections were important during low cholinergic states. Our results demonstrate that short-term plasticity is a critical property of the network that enhances its robustness. Furthermore, simulations suggested that the low and high cholinergic states can each produce runaway excitation through unique mechanisms and different pathologies. Future studies aimed at elucidating the circuit mechanisms of epilepsy could benefit from considering the two modulatory states separately.

Erratum: Recurrent synapses and circuits in the CA3 region of the hippocampus: an associative network

We should like to correct the name for one of the authors. Maria T.Teleñczuk should be corrected to Maria T.Teleńczuk. ACKNOWLEDGEMENTS. Maria T.Teleńczuk would also like to acknowledge support from the Ecole des Neurosciences and the DIM Cerveau et Pensée (Paris, Ile de France).

Observations of dynamical behavior in a stochastic Wilson-Cowan population with plasticity

Understanding network connectivity and its role in brain activity is an arduous task. Complicating matters further is the introduction of synaptic plasticity rules. Observations using a mean-field perspective [1] are by their nature incomplete so, here, a stochastic model, which includes fluctuations, has been employed. This analysis shows that two types of network connections, driven by plasticity, exhibit oscillatory behavior signaled by a flipping between Up and Down states. Fluctuations in each state in both setups display power law-like avalanche distributions. This study, employing a stochastic algorithm [2] used previously in a population-based model [3], introduces plasticity, according to a modified version of [4], into both an E ® E and I ® E network (Figure 1A). The former network includes plastic excitatory, anti-Hebbian synapses, connecting the populations, while the latter contains plastic inhibitory Hebbian synapses. Both networks incorporate a constant recurrent excitatory synapse. Dynamically, each network undergoes oscillations of relaxation type (Figure 1B) with fluctuations whose avalanche distributions look like power laws (Figure 1C). Conclusions

NMDA and GABAB(KIR) Conductances: The “Perfect Couple” for Bistability

Networks that produce persistent firing in response to novel input patterns are thought to be important in working memory and other information storage functions. One possible mechanism for maintaining persistent firing is dendritic voltage bistability in which the depolarized state depends on the voltage dependence of the NMDA conductance at recurrent synapses. In previous models, the hyperpolarized state is dependent on voltage-independent conductances, including GABA(A). The interplay of these conductances leads to bistability, but its robustness is limited by the fact that the conductance ratio must be within a narrow range. The GABA(B) component of inhibitory transmission was not considered in previous analyses. Here, we show that the voltage dependence of the inwardly rectifying potassium (KIR) conductance activated by GABA(B) receptors adds substantial robustness to network simulations of bistability and the persistent firing that it underlies. The hyperpolarized state is robust because, at hyperpolarized potentials, the GABA(B)/KIR conductance is high and the NMDA conductance is low; the depolarized state is robust because, at depolarized potentials, the NMDA conductance is high and the GABA(B)/KIR conductance is low. Our results suggest that this complementary voltage dependence of GABA(B)/KIR and NMDA conductances makes them a "perfect couple" for producing voltage bistability.

Depth-Dependent Temporal Response Properties in Core Auditory Cortex

The computational role of cortical layers within auditory cortex has proven difficult to establish. One hypothesis is that interlaminar cortical processing might be dedicated to analyzing temporal properties of sounds; if so, then there should be systematic depth-dependent changes in cortical sensitivity to the temporal context in which a stimulus occurs. We recorded neural responses simultaneously across cortical depth in primary auditory cortex and anterior auditory field of CBA/Ca mice, and found systematic depth dependencies in responses to second-and-later noise bursts in slow (1-10 bursts/s) trains of noise bursts. At all depths, responses to noise bursts within a train usually decreased with increasing train rate; however, the rolloff with increasing train rate occurred at faster rates in more superficial layers. Moreover, in some recordings from mid-to-superficial layers, responses to noise bursts within a 3-4 bursts/s train were stronger than responses to noise bursts in slower trains. This non-monotonicity with train rate was especially pronounced in more superficial layers of the anterior auditory field, where responses to noise bursts within the context of a slow train were sometimes even stronger than responses to the noise burst at train onset. These findings may reflect depth dependence in suppression and recovery of cortical activity following a stimulus, which we suggest could arise from laminar differences in synaptic depression at feedforward and recurrent synapses.

Network recruitment to coherent oscillations in a hippocampal computer model

Coherent neural oscillations represent transient synchronization of local neuronal populations in both normal and pathological brain activity. These oscillations occur at or above gamma frequencies (>30 Hz) and often are propagated to neighboring tissue under circumstances that are both normal and abnormal, such as gamma binding or seizures. The mechanisms that generate and propagate these oscillations are poorly understood. In the present study we demonstrate, via a detailed computational model, a mechanism whereby physiological noise and coupling initiate oscillations and then recruit neighboring tissue, in a manner well described by a combination of stochastic resonance and coherence resonance. We develop a novel statistical method to quantify recruitment using several measures of network synchrony. This measurement demonstrates that oscillations spread via preexisting network connections such as interneuronal connections, recurrent synapses, and gap junctions, provided that neighboring cells also receive sufficient inputs in the form of random synaptic noise. "Epileptic" high-frequency oscillations (HFOs), produced by pathologies such as increased synaptic activity and recurrent connections, were superior at recruiting neighboring tissue. "Normal" HFOs, associated with fast firing of inhibitory cells and sparse pyramidal cell firing, tended to suppress surrounding cells and showed very limited ability to recruit. These findings point to synaptic noise and physiological coupling as important targets for understanding the generation and propagation of both normal and pathological HFOs, suggesting potential new diagnostic and therapeutic approaches to human disorders such as epilepsy.

Membrane and synaptic properties of pyramidal neurons in the anterior olfactory nucleus

The anterior olfactory nucleus (AON) is positioned to coordinate activity between the piriform cortex and olfactory bulbs, yet the physiology of AON principal neurons has been little explored. Here, we examined the membrane properties and excitatory synapses of AON principal neurons in brain slices of PND22-28 mice and compared their properties to principal cells in other olfactory cortical areas. AON principal neurons had firing rates, spike rate adaptation, spike widths, and I-V relationships that were generally similar to pyramidal neurons in piriform cortex, and typical of cerebral cortex, consistent with a role for AON in cortical processing. Principal neurons in AON had more hyperpolarized action potential thresholds, smaller afterhyperpolarizations, and tended to fire doublets of action potentials on depolarization compared with ventral anterior piriform cortex and the adjacent epileptogenic region preendopiriform nucleus (pEN). Thus, AON pyramidal neurons have enhanced membrane excitability compared with surrounding subregions. Interestingly, principal neurons in pEN were the least excitable, as measured by a larger input conductance, lower firing rates, and more inward rectification. Afferent and recurrent excitatory synapses onto AON pyramidal neurons had small amplitudes, paired pulse facilitation at afferent synapses, and GABA(B) modulation at recurrent synapses, a pattern similar to piriform cortex. The enhanced membrane excitability and recurrent synaptic excitation within the AON, together with its widespread outputs, suggest that the AON can boost and distribute activity in feedforward and feedback circuits throughout the olfactory system.

Spatiotemporal properties of the action potential propagation in the mouse visual cortical slice analyzed by calcium imaging.

The calcium ion (Ca2+) is an important messenger for signal transduction, and the intracellular Ca2+ concentration ([Ca2+]i) changes in response to an excitation of the cell. To reveal the spatiotemporal properties of the propagation of an excitatory signal with action potentials in the primary visual cortical circuit, we conducted a Ca2+ imaging study on slices of the mouse visual cortex. Electrical stimulation of layer 4 evoked [Ca2+]i transients around the stimulus electrode. Subsequently, the high [Ca2+]i region mainly propagated perpendicular to the cortical layer (vertical propagation), with horizontal propagation being restricted. When the excitatory synaptic transmission was blocked, only weak and concentric [Ca2+]i transients were observed. When the action potential was blocked, the [Ca2+]i transients disappeared almost completely. These results suggested that the action potential contributed to the induction of the [Ca2+]i transients, and that excitatory synaptic connections were involved in the propagation of the high [Ca2+]i region in the primary visual cortical circuit. To elucidate the involvement of inhibitory synaptic connections in signal propagation in the primary visual cortex, the GABAA receptor inhibitor bicuculline was applied. In this case, the evoked signal propagated from layer 4 to the entire field of view, and the prolonged [Ca2+]i transients were observed compared with the control condition. Our results suggest that excitatory neurons are widely connected to each other over the entire primary visual cortex with recurrent synapses, and inhibitory neurons play a fundamental role in the organization of functional sub-networks by restricting the propagation of excitation signals.

Emergence of network structure due to spike-timing-dependent plasticity in recurrent neuronal networks V: self-organization schemes and weight dependence

Spike-timing-dependent plasticity (STDP) determines the evolution of the synaptic weights according to their pre- and post-synaptic activity, which in turn changes the neuronal activity on a (much) slower time scale. This paper examines the effect of STDP in a recurrently connected network stimulated by external pools of input spike trains, where both input and recurrent synapses are plastic. Our previously developed theoretical framework is extended to incorporate weight-dependent STDP and dendritic delays. The weight dynamics is determined by an interplay between the neuronal activation mechanisms, the input spike-time correlations, and the learning parameters. For the case of two external input pools, the resulting learning scheme can exhibit a symmetry breaking of the input connections such that two neuronal groups emerge, each specialized to one input pool only. In addition, we show how the recurrent connections within each neuronal group can be strengthened by STDP at the expense of those between the two groups. This neuronal self-organization can be seen as a basic dynamical ingredient for the emergence of neuronal maps induced by activity-dependent plasticity.

GluK1 inhibits calcium dependent and independent transmitter release at associational/commissural synapses in area CA3 of the hippocampus

CA3 pyramidal cells receive three main excitatory inputs: the first one is the mossy fiber input, synapsing mainly on the proximal apical dendrites. Second, entorhinal cortex cells form excitatory connections with CA3 pyramidal cells via the perforant path in the stratum lacunosum moleculare. The third input involves the ipsi-and contralateral connections, termed the associational/commissural (A/C) pathway terminating in the stratum radiatum of CA3, thus forming a feedback loop within this region. Since this excitatory recurrent synapse makes the CA3 region extremely prone to seizure development, understanding the regulation of synaptic strength of this connection is of crucial interest. Several studies suggest that kainate receptors (KAR) play a role in the regulation of synaptic strength. Our aim was to characterize the influence of KAR on A/C synaptic transmission: application of ATPA, a selective agonist of the GluK1 KAR, depressed the amplitude fEPSP without affecting the size of the fiber volley. Blockade of GABA receptors had no influence on this effect, arguing against the influence of interneuronal KARs. Pharmacological and genetic deletion studies could show that this effect was selectively due to GluK1 receptor activation. Several lines of evidence, such as PPF changes, coefficient of variance-analysis and glutamate uncaging experiments strongly argue for a presynaptic locus of suppression. This is accompanied by an ATPA-mediated reduction in Ca(2+) influx at excitatory synaptic terminals, which is most likely mediated by a G-Protein dependent mechanism, as suggested by application of pertussis toxin. Finally, analysis of miniature EPSCs in the presence and absence of extracellular Ca(2+) suggest that presynaptic KAR can also reduce transmitter release downstream and therefore independent of Ca(2+) influx.

Ablation of NMDA Receptors Enhances the Excitability of Hippocampal CA3 Neurons

Synchronized discharges in the hippocampal CA3 recurrent network are supposed to underlie network oscillations, memory formation and seizure generation. In the hippocampal CA3 network, NMDA receptors are abundant at the recurrent synapses but scarce at the mossy fiber synapses. We generated mutant mice in which NMDA receptors were abolished in hippocampal CA3 pyramidal neurons by postnatal day 14. The histological and cytological organizations of the hippocampal CA3 region were indistinguishable between control and mutant mice. We found that mutant mice lacking NMDA receptors selectively in CA3 pyramidal neurons became more susceptible to kainate-induced seizures. Consistently, mutant mice showed characteristic large EEG spikes associated with multiple unit activities (MUA), suggesting enhanced synchronous firing of CA3 neurons. The electrophysiological balance between fast excitatory and inhibitory synaptic transmission was comparable between control and mutant pyramidal neurons in the hippocampal CA3 region, while the NMDA receptor-slow AHP coupling was diminished in the mutant neurons. In the adult brain, inducible ablation of NMDA receptors in the hippocampal CA3 region by the viral expression vector for Cre recombinase also induced similar large EEG spikes. Furthermore, pharmacological blockade of CA3 NMDA receptors enhanced the susceptibility to kainate-induced seizures. These results raise an intriguing possibility that hippocampal CA3 NMDA receptors may suppress the excitability of the recurrent network as a whole in vivo by restricting synchronous firing of CA3 neurons.

Factors defining a pacemaker region for synchrony in the hippocampus

Synchronous activities of neuronal populations are often initiated in a pacemaker region and spread to recruit other regions. Here we examine factors that define a pacemaker site. The CA3a region acts as the pacemaker for disinhibition induced synchrony in guinea pig hippocampal slices and CA3b is a follower region. We found CA3a pyramidal cells were more excitable and fired in bursts more frequently than CA3b cells. CA3a cells had more complex dendritic arbors than CA3b cells especially in zones targetted by recurrent synapses. The product of the density of pyramidal cell axon terminals and dendritic lengths in innervated zones predicted a higher recurrent synaptic connectivity in the CA3a than in the CA3b region. We show that some CA3a cells but few CA3b cells behave as pacemaker cells by firing early during population events and by recruiting follower cells to fire. With a greater excitability and enhanced synaptic connectivity these CA3a cells may also possess initiating functions for other hippocampal ensemble activities initiated in this region.

Activity-Dependent Depression of Local Excitatory Connections in the CA1 Region of Mouse Hippocampus

The existence of recurrent excitatory synapses between pyramidal cells in the hippocampal CA1 region has been known for some time yet little is known about activity-dependent forms of plasticity at these synapses. Here we demonstrate that under certain experimental conditions, Schaffer collateral/commissural fiber stimulation can elicit robust polysynaptic excitatory postsynaptic potentials due to recurrent synaptic inputs onto CA1 pyramidal cells. In contrast to CA3 pyramidal cell inputs, recurrent synapses onto CA1 pyramidal cells exhibited robust paired-pulse depression and a sustained, but rapidly reversible, depression in response to low-frequency trains of Schaffer collateral fiber stimulation. Blocking GABA(B) receptors abolished paired-pulse depression but had little effect on low-frequency stimulation (LFS)-induced depression. Instead, LFS-induced depression was significantly attenuated by an inhibitor of A1 type adenosine receptors. Blocking the postsynaptic effects of GABA(B) and A1 receptor activation on CA1 pyramidal cell excitability with an inhibitor of G-protein-activated inwardly rectifying potassium channels had no effect on either paired-pulse depression or LFS-induced depression. Thus activation of presynaptic GABA(B) and adenosine receptors appears to have an important role in activity-dependent depression at recurrent synapses. Together, our results indicate that CA3-CA1 and CA1-CA1 synapses exhibit strikingly different forms of short-term synaptic plasticity and suggest that activity-dependent changes in recurrent synaptic transmission can transform the CA1 region from a sparsely connected recurrent network into a predominantly feedforward circuit.

Adaptivity of Tuning Functions in a Generic Recurrent Network Model of a Cortical Hypercolumn

The representation of orientation information in the adult visual cortex is plastic as exemplified by phenomena such as perceptual learning or attention. Although these phenomena operate on different time scales and give rise to different changes in the response properties of neurons, both lead to an improvement in visual discrimination or detection tasks. If, however, optimal performance is indeed the goal, the question arises as to why the changes in neuronal response properties are so different. Here, we hypothesize that these differences arise naturally if optimal performance is achieved by means of different mechanisms. To evaluate this hypothesis, we set up a recurrent network model of a visual cortical hypercolumn and asked how each of four different parameter sets (strength of afferent and recurrent synapses, neuronal gains, and additive background inputs) must be changed to optimally improve the encoding accuracy of a particular set of visual stimuli. We find that the predicted changes in the population responses and the tuning functions were different for each set of parameters, hence were strongly dependent on the plasticity mechanism that was operative. An optimal change in the strength of the recurrent connections, for example, led to changes in the response properties that are similar to the changes observed in perceptual learning experiments. An optimal change in the neuronal gains led to changes mimicking neural effects of attention. Assuming the validity of the optimal encoding hypothesis, these model predictions can be used to disentangle the mechanisms of perceptual learning, attention, and other adaptation phenomena.