Dynamics and plasticity in developing neuronal networks in vitro

Engineered cortical microcircuits for investigations of neuroplasticity

Recent advances in neural engineering have opened new ways to investigate the impact of topology on neural network function. Leveraging microfluidic technologies, it is possible to establish modular circuit motifs that promote both segregation and integration of information processing in the engineered neural networks, similar to those observed in vivo. However, the impact of the underlying topologies on network dynamics and response to pathological perturbation remains largely unresolved. In this work, we demonstrate the utilization of microfluidic platforms with 12 interconnected nodes to structure modular, cortical engineered neural networks. By implementing geometrical constraints inspired by a Tesla valve within the connecting microtunnels, we additionally exert control over the direction of axonal outgrowth between the nodes. Interfacing these platforms with nanoporous microelectrode arrays reveals that the resulting laminar cortical networks exhibit pronounced segregated and integrated functional dynamics across layers, mirroring key elements of the feedforward, hierarchical information processing observed in the neocortex. The multi-nodal configuration also facilitates selective perturbation of individual nodes within the networks. To illustrate this, we induced hypoxia, a key factor in the pathogenesis of various neurological disorders, in well-connected nodes within the networks. Our findings demonstrate that such perturbations induce ablation of information flow across the hypoxic node, while enabling the study of plasticity and information processing adaptations in neighboring nodes and neural communication pathways. In summary, our presented model system recapitulates fundamental attributes of the microcircuit organization of neocortical neural networks, rendering it highly pertinent for preclinical neuroscience research. This model system holds promise for yielding new insights into the development, topological organization, and neuroplasticity mechanisms of the neocortex across the micro- and mesoscale level, in both healthy and pathological conditions.

Long-termin vitroculture of 3D brain tissue model based on chitosan thermogel

Methods for studying brain function and disease heavily rely onin vivoanimal models,ex-vivotissue slices, and 2D cell culture platforms. These methods all have limitations that significantly impact the clinical translatability of results. Consequently, models able to better recapitulate some aspects ofin vivohuman brain are needed as additional preclinical tools. In this context, 3D hydrogel-basedin vitromodels of the brain are considered promising tools. To create a 3D brain-on-a-chip model, a hydrogel capable of sustaining neuronal maturation over extended culture periods is required. Among biopolymeric hydrogels, chitosan-β-glycerophosphate (CHITO-β-GP) thermogels have demonstrated their versatility and applicability in the biomedical field over the years. In this study, we investigated the ability of this thermogel to encapsulate neuronal cells and support the functional maturation of a 3D neuronal network in long-term cultures. To the best of our knowledge, we demonstrated for the first time that CHITO-β-GP thermogel possesses optimal characteristics for promoting neuronal growth and the development of an electrophysiologically functional neuronal network derived from both primary rat neurons and neurons differentiated from human induced pluripotent stem cells (h-iPSCs) co-cultured with astrocytes. Specifically, two different formulations were firstly characterized by rheological, mechanical and injectability tests. Primary nervous cells and neurons differentiated from h-iPSCs were embedded into the two thermogel formulations. The 3D cultures were then deeply characterized by immunocytochemistry, confocal microscopy, and electrophysiological recordings, employing both 2D and 3D micro-electrode arrays. The thermogels supported the long-term culture of neuronal networks for up to 100 d. In conclusion, CHITO-β-GP thermogels exhibit excellent mechanical properties, stability over time under culture conditions, and bioactivity toward nervous cells. Therefore, they are excellent candidates as artificial extracellular matrices in brain-on-a-chip models, with applications in neurodegenerative disease modeling, drug screening, and neurotoxicity evaluation.

DualSort: online spike sorting with a running neural network

Objective.Spike sorting, i.e. the detection and separation of measured action potentials from different extracellularly recorded neurons, remains one of the bottlenecks in deciphering the brain. In recent years, the application of neural networks (NNs) for spike sorting has garnered significant attention. Most methods focus on specific sub-problems within the conventional spike sorting pipeline, such as spike detection or feature extraction, and attempt to solve them with complex network architectures. This paper presents DualSort, a simple NN that gets combined with downstream post-processing for real-time spike sorting. It shows high efficiency, low complexity, and requires a comparatively small amount of human interaction.Approach.Synthetic and experimentally obtained extracellular single-channel recordings were utilized to train and evaluate the proposed NN. For training, spike waveforms were labeled with respect to their associated neuron and position in the signal, allowing the detection and categorization of spikes in unison. DualSort classifies a single spike multiple times in succession, as it runs over the signal in a step-by-step manner and uses a post-processing algorithm that transmits the network output into spike trains. Main results.With the used datasets, DualSort was able to detect and distinguish different spike waveforms and separate them from background activity. The post-processing algorithm significantly strengthened the overall performance of the model, making the system more robust as a whole. Although DualSort is an end-to-end solution that efficiently transforms filtered signals into spike trains, it competes with contemporary state-of-the-art technologies that exclusively target single sub-problems in the conventional spike sorting pipeline.Significance.This work demonstrates that even under high noise levels, complex NNs are not necessary by any means to achieve high performance in spike detection and sorting. The utilization of data augmentation on a limited quantity of spikes could substantially decrease hand-labeling compared to other studies. Furthermore, the proposed framework can be utilized without human interaction when combined with an unsupervised technique that provides pseudo labels for DualSort. Due to the low complexity of our network, it works efficiently and enables real-time processing on basic hardware. The proposed approach is not limited to spike sorting, as it may also be used to process different signals, such as electroencephalogram (EEG), which needs to be investigated in future research.

Lysosomal alterations and decreased electrophysiological activity in CLN3 disease patient-derived cortical neurons

CLN3 disease is a lysosomal storage disorder associated with fatal neurodegeneration that is caused by mutations in CLN3, with most affected individuals carrying at least one allele with a 966 bp deletion. Using CRISPR/Cas9, we corrected the 966 bp deletion mutation in human induced pluripotent stem cells (iPSCs) of a compound heterozygous patient (CLN3 Δ 966 bp and E295K). We differentiated these isogenic iPSCs, and iPSCs from an unrelated healthy control donor, to neurons and identified disease-related changes relating to protein synthesis, trafficking and degradation, and in neuronal activity, which were not apparent in CLN3-corrected or healthy control neurons. CLN3 neurons showed numerous membrane-bound vacuoles containing diverse storage material and hyperglycosylation of the lysosomal LAMP1 protein. Proteomic analysis showed increase in lysosomal-related proteins and many ribosomal subunit proteins in CLN3 neurons, accompanied by downregulation of proteins related to axon guidance and endocytosis. CLN3 neurons also had lower electrophysical activity as recorded using microelectrode arrays. These data implicate inter-related pathways in protein homeostasis and neurite arborization as contributing to CLN3 disease, and which could be potential targets for therapy.

An automated method for precise axon reconstruction from recordings of high-density micro-electrode arrays

Objective:Neurons communicate with each other by sending action potentials (APs) through their axons. The velocity of axonal signal propagation describes how fast electrical APs can travel. This velocity can be affected in a human brain by several pathologies, including multiple sclerosis, traumatic brain injury and channelopathies. High-density microelectrode arrays (HD-MEAs) provide unprecedented spatio-temporal resolution to extracellularly record neural electrical activity. The high density of the recording electrodes enables to image the activity of individual neurons down to subcellular resolution, which includes the propagation of axonal signals. However, axon reconstruction, to date, mainly relies on manual approaches to select the electrodes and channels that seemingly record the signals along a specific axon, while an automated approach to track multiple axonal branches in extracellular action-potential recordings is still missing.Approach:In this article, we propose a fully automated approach to reconstruct axons from extracellular electrical-potential landscapes, so-called 'electrical footprints' of neurons. After an initial electrode and channel selection, the proposed method first constructs a graph based on the voltage signal amplitudes and latencies. Then, the graph is interrogated to extract possible axonal branches. Finally, the axonal branches are pruned, and axonal action-potential propagation velocities are computed.Main results:We first validate our method using simulated data from detailed reconstructions of neurons, showing that our approach is capable of accurately reconstructing axonal branches. We then apply the reconstruction algorithm to experimental recordings of HD-MEAs and show that it can be used to determine axonal morphologies and signal-propagation velocities at high throughput.Significance:We introduce a fully automated method to reconstruct axonal branches and estimate axonal action-potential propagation velocities using HD-MEA recordings. Our method yields highly reliable and reproducible velocity estimations, which constitute an important electrophysiological feature of neuronal preparations.

Functionalized Nanocellulose Drives Neural Stem Cells toward Neuronal Differentiation

Transplantation of differentiated and fully functional neurons may be a better therapeutic option for the cure of neurodegenerative disorders and brain injuries than direct grafting of neural stem cells (NSCs) that are potentially tumorigenic. However, the differentiation of NSCs into a large population of neurons has been a challenge. Nanomaterials have been widely used as substrates to manipulate cell behavior due to their nano-size, excellent physicochemical properties, ease of synthesis, and versatility in surface functionalization. Nanomaterial-based scaffolds and synthetic polymers have been fabricated with topology resembling the micro-environment of the extracellular matrix. Nanocellulose materials are gaining attention because of their availability, biocompatibility, biodegradability and bioactivity, and affordable cost. We evaluated the role of nanocellulose with different linkage and surface features in promoting neuronal differentiation. Nanocellulose coupled with lysine molecules (CNC-Lys) provided positive charges that helped the cells to attach. Embryonic rat NSCs were differentiated on the CNC-Lys surface for up to three weeks. By the end of the three weeks of in vitro culture, 87% of the cells had attached to the CNC-Lys surface and more than half of the NSCs had differentiated into functional neurons, expressing endogenous glutamate, generating electrical activity and action potentials recorded by the multi-electrode array.

Early prediction of developing spontaneous activity in cultured neuronal networks

Synchronization and bursting activity are intrinsic electrophysiological properties of in vivo and in vitro neural networks. During early development, cortical cultures exhibit a wide repertoire of synchronous bursting dynamics whose characterization may help to understand the parameters governing the transition from immature to mature networks. Here we used machine learning techniques to characterize and predict the developing spontaneous activity in mouse cortical neurons on microelectrode arrays (MEAs) during the first three weeks in vitro. Network activity at three stages of early development was defined by 18 electrophysiological features of spikes, bursts, synchrony, and connectivity. The variability of neuronal network activity during early development was investigated by applying k-means and self-organizing map (SOM) clustering analysis to features of bursts and synchrony. These electrophysiological features were predicted at the third week in vitro with high accuracy from those at earlier times using three machine learning models: Multivariate Adaptive Regression Splines, Support Vector Machines, and Random Forest. Our results indicate that initial patterns of electrical activity during the first week in vitro may already predetermine the final development of the neuronal network activity. The methodological approach used here may be applied to explore the biological mechanisms underlying the complex dynamics of spontaneous activity in developing neuronal cultures.

Differential Excitability of PV and SST Neurons Results in Distinct Functional Roles in Inhibition Stabilization of Up States

Up states are the best studied example of an emergent neural dynamic regime. Computational models based on a single class of inhibitory neurons indicate that Up states reflect bistable dynamic systems in which positive feedback is stabilized by strong inhibition and predict a paradoxical effect in which increased drive to inhibitory neurons results in decreased inhibitory activity. To date, however, computational models have not incorporated empirically defined properties of parvalbumin (PV) and somatostatin (SST) neurons. Here we first experimentally characterized the frequency-current (F-I) curves of pyramidal (Pyr), PV, and SST neurons from mice of either sex, and confirmed a sharp difference between the threshold and slopes of PV and SST neurons. The empirically defined F-I curves were incorporated into a three-population computational model that simulated the empirically derived firing rates of pyramidal, PV, and SST neurons. Simulations revealed that the intrinsic properties were sufficient to predict that PV neurons are primarily responsible for generating the nontrivial fixed points representing Up states. Simulations and analytical methods demonstrated that while the paradoxical effect is not obligatory in a model with two classes of inhibitory neurons, it is present in most regimes. Finally, experimental tests validated predictions of the model that the Pyr ↔ PV inhibitory loop is stronger than the Pyr ↔ SST loop.SIGNIFICANCE STATEMENT Many cortical computations, such as working memory, rely on the local recurrent excitatory connections that define cortical circuit motifs. Up states are among the best studied examples of neural dynamic regimes that rely on recurrent excitatory excitation. However, this positive feedback must be held in check by inhibition. To address the relative contribution of PV and SST neurons, we characterized the intrinsic input-output differences between these classes of inhibitory neurons and, using experimental and theoretical methods, show that the higher threshold and gain of PV leads to a dominant role in network stabilization.

Partial information decomposition reveals that synergistic neural integration is greater downstream of recurrent information flow in organotypic cortical cultures

The directionality of network information flow dictates how networks process information. A central component of information processing in both biological and artificial neural networks is their ability to perform synergistic integration–a type of computation. We established previously that synergistic integration varies directly with the strength of feedforward information flow. However, the relationships between both recurrent and feedback information flow and synergistic integration remain unknown. To address this, we analyzed the spiking activity of hundreds of neurons in organotypic cultures of mouse cortex. We asked how empirically observed synergistic integration–determined from partial information decomposition–varied with local functional network structure that was categorized into motifs with varying recurrent and feedback information flow. We found that synergistic integration was elevated in motifs with greater recurrent information flow beyond that expected from the local feedforward information flow. Feedback information flow was interrelated with feedforward information flow and was associated with decreased synergistic integration. Our results indicate that synergistic integration is distinctly influenced by the directionality of local information flow.

Networks compute information. That is, they modify inputs to generate distinct outputs. These computations are an important component of network information processing. Knowing how the routing of information in a network influences computation is therefore crucial. Here we asked how a key form of computation—synergistic integration—is related to the direction of local information flow in networks of spiking cortical neurons. Specifically, we asked how information flow between input neurons (i.e., recurrent information flow) and information flow from output neurons to input neurons (i.e., feedback information flow) was related to the amount of synergistic integration performed by output neurons. We found that greater synergistic integration occurred where there was more recurrent information flow. And, lesser synergistic integration occurred where there was more feedback information flow relative to feedforward information flow. These results show that computation, in the form of synergistic integration, is distinctly influenced by the directionality of local information flow. Such work is valuable for predicting where and how network computation occurs and for designing networks with desired computational abilities.

Consolidation of memory traces in cultured cortical networks requires low cholinergic tone, synchronized activity and high network excitability

In systems consolidation, encoded memories are replayed by the hippocampus during slow-wave sleep (SWS), and permanently stored in the neocortex. Declarative memory consolidation is believed to benefit from the oscillatory rhythms and low cholinergic tone observed in this sleep stage, but underlying mechanisms remain unclear. To clarify the role of cholinergic modulation and synchronized activity in memory consolidation, we applied repeated electrical stimulation in mature cultures of dissociated rat cortical neurons with high or low cholinergic tone, mimicking the cue replay observed during systems consolidation under distinct cholinergic concentrations. In the absence of cholinergic input, these cultures display activity patterns hallmarked by network bursts, synchronized events reminiscent of the low frequency oscillations observed during SWS. They display stable activity and connectivity, which mutually interact and achieve an equilibrium. Electrical stimulation reforms the equilibrium to include the stimulus response, a phenomenon interpreted as memory trace formation. Without cholinergic input, activity was burst-dominated. First application of a stimulus induced significant connectivity changes, while subsequent repetition no longer affected connectivity. Presenting a second stimulus at a different electrode had the same effect, whereas returning to the initial stimuli did not induce further connectivity alterations, indicating that the second stimulus did not erase the 'memory trace' of the first. Distinctively, cultures with high cholinergic tone displayed reduced network excitability and dispersed firing, and electrical stimulation did not induce significant connectivity changes. We conclude that low cholinergic tone facilitates memory formation and consolidation, possibly through enhanced network excitability. Network bursts or SWS oscillations may merely reflect high network excitability.

Deficits in coordinated neuronal activity and network topology are striatal hallmarks in Huntington's disease

Background

Network alterations underlying neurodegenerative diseases often precede symptoms and functional deficits. Thus, their early identification is central for improved prognosis. In Huntington’s disease (HD), the cortico-striatal networks, involved in motor function processing, are the most compromised neural substrate. However, whether the network alterations are intrinsic of the striatum or the cortex is not fully understood.

Results

In order to identify early HD neural deficits, we characterized neuronal ensemble calcium activity and network topology of HD striatal and cortical cultures. We used large-scale calcium imaging combined with activity-based network inference analysis. We extracted collective activity events and inferred the topology of the neuronal network in cortical and striatal primary cultures from wild-type and R6/1 mouse model of HD. Striatal, but not cortical, HD networks displayed lower activity and a lessened ability to integrate information. GABA_A receptor blockade in healthy and HD striatal cultures generated similar coordinated ensemble activity and network topology, highlighting that the excitatory component of striatal system is spared in HD. Conversely, NMDA receptor activation increased individual neuronal activity while coordinated activity became highly variable and undefined. Interestingly, by boosting NMDA activity, we rectified striatal HD network alterations.

Conclusions

Overall, our integrative approach highlights striatal defective network integration capacity as a major contributor of basal ganglia dysfunction in HD and suggests that increased excitatory drive may serve as a potential intervention. In addition, our work provides a valuable tool to evaluate in vitro network recovery after treatment intervention in basal ganglia disorders.

Bilateral enucleation at birth modifies calcium spike amplitude, but not frequency, in neurons of the somatosensory thalamus and cortex: Implications for developmental cross-modal plasticity

Bilateral eye enucleation at birth (BE) leads to an expansion of the primary somatosensory cortex (S1) in rat pups. Although increased growth of the somatosensory thalamo-cortical afferents (STCAs) in part explains S1 expansion, timing mechanisms governing S1 formation are also involved. In this work, we begin the search of a developmental clock by intending to document the existence of putative clock neurons in the somatosensory thalamus (VPM) and S1 based upon changes of spontaneous spike amplitude; a biophysical property sensitive to circadian regulation; the latter known to be shifted by enucleation. In addition, we also evaluated whether STCAs growth rate and segregation timing were modified, as parameters the clock might time. We found that spontaneous spike amplitude transiently, but significantly, increased or decreased in VPM and S1 neurons of BE rat pups, respectively, as compared to their control counterparts. The growth rate and segregation timing of STCAs was, however, unaffected by BE. These results support the existence of a developmental clock that ticks differently in the VPM and S1 after BE. This observation, together with the fact that STCAs growth rate and segregation timing is unchanged, suggests that S1 expansion in BE rats may in part be controlled at the cortical level.

Hindbrain V2a Neurons Pattern Rhythmic Activity of Motor Neurons in a Reticulospinal Coculture

As the capacity to isolate distinct neuronal cell types has advanced over the past several decades, new two- and three-dimensional in vitro models of the interactions between different brain regions have expanded our understanding of human neurobiology and the origins of disease. These cultures develop distinctive patterns of activity, but the extent that these patterns are determined by the molecular identity of individual cell types versus the specific pattern of network connectivity is unclear. To address the question of how individual cell types interact in vitro, we developed a simplified culture using two excitatory neuronal subtypes known to participate in the in vivo reticulospinal circuit: HB9+ spinal motor neurons and Chx10+ hindbrain V2a neurons. Here, we report the emergence of cell type-specific patterns of activity in culture; on their own, Chx10+ neurons developed regular, synchronized bursts of activity that recruited neurons across the entire culture, whereas HB9+ neuron activity consisted of an irregular pattern. When these two subtypes were cocultured, HB9+ neurons developed synchronized network bursts that were precisely correlated with Chx10+ neuron activity, thereby recreating an aspect of Chx10+ neurons' role in driving motor activity. These bursts were dependent on AMPA receptors. Our results demonstrate that the molecular classification of the neurons comprising in vitro networks is a crucial determinant of their activity. It is therefore possible to improve both the reproducibility and the applicability of in vitro neurobiological and disease models by carefully controlling the constituent mixtures of neuronal subtypes.

Computational Modeling Reveals Frequency Modulation of Calcium-cAMP/PKA Pathway in Dendritic Spines

Dendritic spines are the primary excitatory postsynaptic sites that act as subcompartments of signaling. Ca²⁺ is often the first and most rapid signal in spines. Downstream of calcium, the cyclic adenosine monophosphate (cAMP)/protein kinase A (PKA) pathway plays a critical role in the regulation of spine formation, morphological modifications, and ultimately, learning and memory. Although the dynamics of calcium are reasonably well-studied, calcium-induced cAMP/PKA dynamics, particularly with respect to frequency modulation, are not fully explored. In this study, we present a well-mixed model for the dynamics of calcium-induced cAMP/PKA dynamics in dendritic spines. The model is constrained using experimental observations in the literature. Further, we measured the calcium oscillation frequency in dendritic spines of cultured hippocampal CA1 neurons and used these dynamics as model inputs. Our model predicts that the various steps in this pathway act as frequency modulators for calcium, and the high frequency of calcium input is filtered by adenylyl cyclase 1 and phosphodiesterases in this pathway such that cAMP/PKA only responds to lower frequencies. This prediction has important implications for noise filtering and long-timescale signal transduction in dendritic spines. A companion manuscript presents a three-dimensional spatial model for the same pathway.

Self-organization of modular network architecture by activity-dependent neuronal migration and outgrowth

The spatial distribution of neurons and activity-dependent neurite outgrowth shape long-range interaction, recurrent local connectivity and the modularity in neuronal networks. We investigated how this mesoscale architecture develops by interaction of neurite outgrowth, cell migration and activity in cultured networks of rat cortical neurons and show that simple rules can explain variations of network modularity. In contrast to theoretical studies on activity-dependent outgrowth but consistent with predictions for modular networks, spontaneous activity and the rate of synchronized bursts increased with clustering, whereas peak firing rates in bursts increased in highly interconnected homogeneous networks. As Ca²⁺ influx increased exponentially with increasing network recruitment during bursts, its modulation was highly correlated to peak firing rates. During network maturation, long-term estimates of Ca²⁺ influx showed convergence, even for highly different mesoscale architectures, neurite extent, connectivity, modularity and average activity levels, indicating homeostatic regulation towards a common set-point of Ca²⁺ influx.

Long-Term Activity Dynamics of Single Neurons and Networks

The firing rate of neuronal spiking in vitro and in vivo significantly varies over extended timescales, characterized by long-memory processes and complex statistics, and appears in spontaneous as well as evoked activity upon repeated stimulus presentation. These variations in response features and their statistics, in face of repeated instances of a given physical input, are ubiquitous in all levels of brain-behavior organization. They are expressed in single neuron and network response variability but even appear in variations of subjective percepts or psychophysical choices and have been described as stemming from history-dependent, stochastic, or rate-determined processes.But what are the sources underlying these temporally rich variations in firing rate? Are they determined by interactions of the nervous system as a whole, or do isolated, single neurons or neuronal networks already express these fluctuations independent of higher levels? These questions motivated the application of a method that allows for controlled and specific long-term activation of a single neuron or neuronal network, isolated from higher levels of cortical organization.This chapter highlights the research done in cultured cortical networks to study (1) the inherent non-stationarity of neuronal network activity, (2) single neuron response fluctuations and underlying processes, and (3) the interface layer between network and single cell, the non-stationary efficacy of the ensemble of synapses impinging onto the observed neuron.

Application of Microelectrode Array Approaches to Neurotoxicity Testing and Screening

Neurotoxicity can be defined by the ability of a drug or chemical to alter the physiology, biochemistry, or structure of the nervous system in a manner that may negatively impact the health or function of the individual. Electrophysiological approaches have been utilized to study the mechanisms underlying neurotoxic actions of drugs and chemicals for over 50 years, and in more recent decades, high-throughput patch-clamp approaches have been utilized by the pharmaceutical industry for drug development. The use of microelectrode array recordings to study neural network electrophysiology is a relatively newer approach, with commercially available systems becoming available only in the early 2000s. However, MEAs have been rapidly adopted as a useful approach for neurotoxicity testing. In this chapter, I will review the use of MEA approaches as they have been applied to the field of neurotoxicity testing, especially as they have been applied to the need to screen large numbers of chemicals for neurotoxicity and developmental neurotoxicity. In addition, I will also identify challenges for the field that when addressed will improve the utility of MEA approaches for toxicity testing.

Decreased spontaneous electrical activity in neuronal networks exposed to radiofrequency 1,800 MHz signals

The rapid development of wireless communications has raised questions about their potential health risks. So far, the only identified biological effects of radiofrequency fields (RF) are known to be caused by heating, but the issue of potential nonthermal biological effects, especially on the central nervous system (CNS), remains open. We previously reported a decrease in the firing and bursting rates of neuronal cultures exposed to a Global System for Mobile (GSM) RF field at 1,800 MHz for 3 min (Moretti D, Garenne A, Haro E, Poulleier de Gannes F, Lagroye I, Lévêque P, Veyret B, Lewis N. Bioelectromagnetics 34: 571-578, 2013). The aim of the present work was to assess the dose-response relationship for this effect and also to identify a potential differential response elicited by pulse-modulated GSM and continuous-wave (CW) RF fields. Spontaneous bursting activity of neuronal cultures from rat embryonic cortices was recorded using 60-electrode multielectrode arrays (MEAs). At 17-28 days in vitro, the neuronal cultures were subjected to 15-min RF exposures, at specific absorption rates (SAR) ranging from 0.01 to 9.2 W/kg. Both GSM and CW signals elicited a clear decrease in bursting rate during the RF exposure phase. This effect became more marked with increasing SAR and lasted even beyond the end of exposure for the highest SAR levels. Moreover, the amplitude of the effect was greater with the GSM signal. Altogether, our experimental findings provide evidence for dose-dependent effects of RF signals on the bursting rate of neuronal cultures and suggest that part of the mechanism is nonthermal. NEW & NOTEWORTHY In this study, we investigated the effects of some radiofrequency (RF) exposure parameters on the electrical activity of neuronal cultures. We detected a clear decrease in bursting activity, dependent on exposure duration. The amplitude of this effect increased with the specific absorption rate (SAR) level and was greater with Global System for Mobile signal than with continuous-wave signal, at the same average SAR. Our experiment provides unique evidence of a decrease in electrical activity of cortical neuronal cultures during RF exposure.

NDDN: A Cloud-Based Neuroinformation Database for Developing Neuronal Networks

Electrical activity of developing dissociated neuronal networks is of immense significance for understanding the general properties of neural information processing and storage. In addition, the complexity and diversity of network activity patterns make them ideal candidates for developing novel computational models and evaluating algorithms. However, there are rare databases which focus on the changing network dynamics during development. Here, we describe the design and implementation of Neuroinformation Database for Developing Networks (NDDN), a repository for electrophysiological data collected from long-term cultured hippocampal networks. The NDDN contains over 15 terabytes of multielectrode array data consisting of 25,380 items collected from 105 culture batches. Metadata including culturing and recording information and stimulation/drug application protocols are linked to each data item. A Matlab toolbox named MEAKit is also provided with the NDDN to ease the analysis of downloaded data items. We expect that NDDN may contribute to both the fields of experimental and computational neuroscience.

Multimed: An Integrated, Multi-Application Platform for the Real-Time Recording and Sub-Millisecond Processing of Biosignals

Enhanced understanding and control of electrophysiology mechanisms are increasingly being hailed as key knowledge in the fields of modern biology and medicine. As more and more excitable cell mechanics are being investigated and exploited, the need for flexible electrophysiology setups becomes apparent. With that aim, we designed Multimed, which is a versatile hardware platform for the real-time recording and processing of biosignals. Digital processing in Multimed is an arrangement of generic processing units from a custom library. These can freely be rearranged to match the needs of the application. Embedded onto a Field Programmable Gate Array (FPGA), these modules utilize full-hardware signal processing to lower processing latency. It achieves constant latency, and sub-millisecond processing and decision-making on 64 channels. The FPGA core processing unit makes Multimed suitable as either a reconfigurable electrophysiology system or a prototyping platform for VLSI implantable medical devices. It is specifically designed for open- and closed-loop experiments and provides consistent feedback rules, well within biological microseconds timeframes. This paper presents the specifications and architecture of the Multimed system, then details the biosignal processing algorithms and their digital implementation. Finally, three applications utilizing Multimed in neuroscience and diabetes research are described. They demonstrate the system’s configurability, its multi-channel, real-time processing, and its feedback control capabilities.

GABA and Gap Junctions in the Development of Synchronized Activity in Human Pluripotent Stem Cell-Derived Neural Networks

The electrical activity of the brain arises from single neurons communicating with each other. However, how single neurons interact during early development to give rise to neural network activity remains poorly understood. We studied the emergence of synchronous neural activity in human pluripotent stem cell (hPSC)-derived neural networks simultaneously on a single-neuron level and network level. The contribution of gamma-aminobutyric acid (GABA) and gap junctions to the development of synchronous activity in hPSC-derived neural networks was studied with GABA agonist and antagonist and by blocking gap junctional communication, respectively. We characterized the dynamics of the network-wide synchrony in hPSC-derived neural networks with high spatial resolution (calcium imaging) and temporal resolution microelectrode array (MEA). We found that the emergence of synchrony correlates with a decrease in very strong GABA excitation. However, the synchronous network was found to consist of a heterogeneous mixture of synchronously active cells with variable responses to GABA, GABA agonists and gap junction blockers. Furthermore, we show how single-cell distributions give rise to the network effect of GABA, GABA agonists and gap junction blockers. Finally, based on our observations, we suggest that the earliest form of synchronous neuronal activity depends on gap junctions and a decrease in GABA induced depolarization but not on GABA_A mediated signaling.

Dynamic Hebbian Cross-Correlation Learning Resolves the Spike Timing Dependent Plasticity Conundrum

Spike Timing-Dependent Plasticity has been found to assume many different forms. The classic STDP curve, with one potentiating and one depressing window, is only one of many possible curves that describe synaptic learning using the STDP mechanism. It has been shown experimentally that STDP curves may contain multiple LTP and LTD windows of variable width, and even inverted windows. The underlying STDP mechanism that is capable of producing such an extensive, and apparently incompatible, range of learning curves is still under investigation. In this paper, it is shown that STDP originates from a combination of two dynamic Hebbian cross-correlations of local activity at the synapse. The correlation of the presynaptic activity with the local postsynaptic activity is a robust and reliable indicator of the discrepancy between the presynaptic neuron and the postsynaptic neuron's activity. The second correlation is between the local postsynaptic activity with dendritic activity which is a good indicator of matching local synaptic and dendritic activity. We show that this simple time-independent learning rule can give rise to many forms of the STDP learning curve. The rule regulates synaptic strength without the need for spike matching or other supervisory learning mechanisms. Local differences in dendritic activity at the synapse greatly affect the cross-correlation difference which determines the relative contributions of different neural activity sources. Dendritic activity due to nearby synapses, action potentials, both forward and back-propagating, as well as inhibitory synapses will dynamically modify the local activity at the synapse, and the resulting STDP learning rule. The dynamic Hebbian learning rule ensures furthermore, that the resulting synaptic strength is dynamically stable, and that interactions between synapses do not result in local instabilities. The rule clearly demonstrates that synapses function as independent localized computational entities, each contributing to the global activity, not in a simply linear fashion, but in a manner that is appropriate to achieve local and global stability of the neuron and the entire dendritic structure.

Progress in automating patch clamp cellular physiology

Patch clamp electrophysiology has transformed research in the life sciences over the last few decades. Since their inception, automatic patch clamp platforms have evolved considerably, demonstrating the capability to address both voltage- and ligand-gated channels, and showing the potential to play a pivotal role in drug discovery and biomedical research. Unfortunately, the cell suspension assays to which early systems were limited cannot recreate biologically relevant cellular environments, or capture higher order aspects of synaptic physiology and network dynamics. In vivo patch clamp electrophysiology has the potential to yield more biologically complex information and be especially useful in reverse engineering the molecular and cellular mechanisms of single-cell and network neuronal computation, while capturing important aspects of human disease mechanisms and possible therapeutic strategies. Unfortunately, it is a difficult procedure with a steep learning curve, which has restricted dissemination of the technique. Luckily, in vivo patch clamp electrophysiology seems particularly amenable to robotic automation. In this review, we document the development of automated patch clamp technology, from early systems based on multi-well plates through to automated planar-array platforms, and modern robotic platforms capable of performing two-photon targeted whole-cell electrophysiological recordings in vivo.

Functional characterization of GABAA receptor-mediated modulation of cortical neuron network activity in microelectrode array recordings

The numerous γ-aminobutyric acid type A receptor (GABA_AR) subtypes are differentially expressed and mediate distinct functions at neuronal level. In this study we have investigated GABA_AR-mediated modulation of the spontaneous activity patterns of primary neuronal networks from murine frontal cortex by characterizing the effects induced by a wide selection of pharmacological tools at a plethora of activity parameters in microelectrode array (MEA) recordings. The basic characteristics of the primary cortical neurons used in the recordings were studied in some detail, and the expression levels of various GABA_AR subunits were investigated by western blotting and RT-qPCR. In the MEA recordings, the pan-GABA_AR agonist muscimol and the GABA_BR agonist baclofen were observed to mediate phenotypically distinct changes in cortical network activity. Selective augmentation of αβγ GABA_AR signaling by diazepam and of δ-containing GABA_AR (δ-GABA_AR) signaling by DS1 produced pronounced changes in the majority of the activity parameters, both drugs mediating similar patterns of activity changes as muscimol. The apparent importance of δ-GABA_AR signaling for network activity was largely corroborated by the effects induced by the functionally selective δ-GABA_AR agonists THIP and Thio-THIP, whereas the δ-GABA_AR selective potentiator DS2 only mediated modest effects on network activity, even when co-applied with low THIP concentrations. Interestingly, diazepam exhibited dramatically right-shifted concentration-response relationships at many of the activity parameters when co-applied with a trace concentration of DS1 compared to when applied alone. In contrast, the potencies and efficacies displayed by DS1 at the networks were not substantially altered by the concomitant presence of diazepam. In conclusion, the holistic nature of the information extractable from the MEA recordings offers interesting insights into the contributions of various GABA_AR subtypes/subgroups to cortical network activity and the putative functional interplay between these receptors in these neurons.

Recurrently connected and localized neuronal communities initiate coordinated spontaneous activity in neuronal networks

Developing neuronal systems intrinsically generate coordinated spontaneous activity that propagates by involving a large number of synchronously firing neurons. In vivo, waves of spikes transiently characterize the activity of developing brain circuits and are fundamental for activity-dependent circuit formation. In vitro, coordinated spontaneous spiking activity, or network bursts (NBs), interleaved within periods of asynchronous spikes emerge during the development of 2D and 3D neuronal cultures. Several studies have investigated this type of activity and its dynamics, but how a neuronal system generates these coordinated events remains unclear. Here, we investigate at a cellular level the generation of network bursts in spontaneously active neuronal cultures by exploiting high-resolution multielectrode array recordings and computational network modelling. Our analysis reveals that NBs are generated in specialized regions of the network (functional neuronal communities) that feature neuronal links with high cross-correlation peak values, sub-millisecond lags and that share very similar structural connectivity motifs providing recurrent interactions. We show that the particular properties of these local structures enable locally amplifying spontaneous asynchronous spikes and that this mechanism can lead to the initiation of NBs. Through the analysis of simulated and experimental data, we also show that AMPA currents drive the coordinated activity, while NMDA and GABA currents are only involved in shaping the dynamics of NBs. Overall, our results suggest that the presence of functional neuronal communities with recurrent local connections allows a neuronal system to generate spontaneous coordinated spiking activity events. As suggested by the rules used for implementing our computational model, such functional communities might naturally emerge during network development by following simple constraints on distance-based connectivity.

Coordinated spontaneous spiking activity is fundamental for the normal formation of brain circuits during development. However, how ensembles of neurons generate these events remains unclear. To address this question, in the present study, we investigated the network properties that might be required to a neuronal system for the generation of these spontaneous waves of spikes. We performed our study on spontaneously active neuronal cell cultures using high-resolution electrical recordings and a computational network model developed to reproduce our experimental data both quantitatively and qualitatively. Through the analysis of both experimental and simulated data, we found that network bursts are initiated in regions of the network, or “functional communities”, characterized by particular local connectivity properties. We also found that these regions can amplify the background asynchronous spiking activity preceding a network burst and, in this way, can give rise to coordinated spiking events.

As a whole, our results suggest the presence of functional communities of neurons in a developing neuronal system that might naturally emerge by following simple constraints on distance-based connectivity. These regions are most likely required for the generation of the spontaneous coordinated activity that can drive activity-dependent circuit formation.

Investigating the impact of electrical stimulation temporal distribution on cortical network responses

Background

The brain is continuously targeted by a wealth of stimuli with complex spatio-temporal patterns and has presumably evolved in order to cope with those inputs in an optimal way. Previous studies investigating the response capabilities of either single neurons or intact sensory systems to external stimulation demonstrated that stimuli temporal distribution is an important, if often overlooked, parameter.

Results

In this study we investigated how cortical networks plated over micro-electrode arrays respond to different stimulation sequences in which inter-pulse intervals followed a 1/f^β distribution, for different values of β ranging from 0 to ∞. Cross-correlation analysis revealed that network activity preferentially synchronizes with external input sequences featuring β closer to 1 and, in any case, never for regular (i.e. fixed-frequency) stimulation sequences. We then tested the interplay between different average stimulation frequencies (based on the intrinsic firing/bursting frequency of the network) for two selected values of β, i.e. 1 (scale free) and ∞ (regular). In general, we observed no preference for stimulation frequencies matching the endogenous rhythms of the network. Moreover, we found that in case of regular stimulation the capability of the network to follow the stimulation sequence was negatively correlated to the absolute stimulation frequency, whereas using scale-free stimulation cross-correlation between input and output sequences was independent from average input frequency.

Conclusions

Our results point out that the preference for a scale-free distribution of the stimuli is observed also at network level and should be taken into account in designing more efficient protocols for neuromodulation purposes.

Electronic supplementary material

The online version of this article (doi:10.1186/s12868-017-0366-z) contains supplementary material, which is available to authorized users.

Extracellularly Recorded Somatic and Neuritic Signal Shapes and Classification Algorithms for High-Density Microelectrode Array Electrophysiology

High-density microelectrode arrays (HDMEA) have been recently introduced to study principles of neural function at high spatial resolution. However, the exact nature of the experimentally observed extracellular action potentials (EAPs) is still incompletely understood. The soma, axon and dendrites of a neuron can all exhibit regenerative action potentials that could be sensed with HDMEA electrodes. Here, we investigate the contribution of distinct neuronal sources of activity in HDMEA recordings from low-density neuronal cultures. We recorded EAPs with HDMEAs having 11,011 electrodes and then fixed and immunostained the cultures with β3-tubulin for high-resolution fluorescence imaging. Immunofluorescence images overlaid with the activity maps showed EAPs both at neuronal somata and distal neurites. Neuritic EAPs had mostly narrow triphasic shapes, consisting of a positive, a pronounced negative peak and a second positive peak. EAPs near somata had wide monophasic or biphasic shapes with a main negative peak, and following optional positive peak. We show that about 86% of EAP recordings consist of somatic spikes, while the remaining 14% represent neuritic spikes. Furthermore, the adaptation of the waveform shape during bursts of these neuritic spikes suggested that they originate from axons, rather than from dendrites. Our study improves the understanding of HDMEA signals and can aid in the identification of the source of EAPs.

Editor's Highlight: Evaluation of a Microelectrode Array-Based Assay for Neural Network Ontogeny Using Training Set Chemicals

Thousands of compounds in the environment have not been characterized for developmental neurotoxicity (DNT) hazard. To address this issue, methods to screen compounds rapidly for DNT hazard evaluation are necessary and are being developed for key neurodevelopmental processes. In order to develop an assay for network formation, this study evaluated effects of a training set of chemicals on network ontogeny by measuring spontaneous electrical activity in neural networks grown on microelectrode arrays (MEAs). Rat (0-24 h old) primary cortical cells were plated in 48 well-MEA plates and exposed to 6 compounds: acetaminophen, bisindolylmaleimide-1 (Bis-1), domoic acid, mevastatin, sodium orthovanadate, and loperamide for a period of 12 days. Spontaneous network activity was recorded on days 2, 5, 7, 9, and 12 and viability was assessed using the Cell Titer Blue assay on day 12. Network activity (e.g. mean firing rate [MFR], burst rate [BR], etc), increased between days 5 and 12. Random Forest analysis indicated that across all compounds and times, temporal correlation of firing patterns (r), MFR, BR, number of active electrodes and % of spikes in a burst were the most influential parameters in separating control from treated wells. All compounds except acetaminophen (≤ 30 µM) caused concentration-related effects on one or more of these parameters. Domoic acid and sodium orthovanadate altered several of these parameters in the absence of cytotoxicity. Although cytotoxicity was observed with Bis1, mevastatin, and loperamide, some parameters were affected by these compounds at concentrations below those resulting in cytotoxicity. These results demonstrate that this assay may be suitable for screening of compounds for DNT hazard identification.

Feed-Forward Propagation of Temporal and Rate Information between Cortical Populations during Coherent Activation in Engineered In Vitro Networks

Transient propagation of information across neuronal assembles is thought to underlie many cognitive processes. However, the nature of the neural code that is embedded within these transmissions remains uncertain. Much of our understanding of how information is transmitted among these assemblies has been derived from computational models. While these models have been instrumental in understanding these processes they often make simplifying assumptions about the biophysical properties of neurons that may influence the nature and properties expressed. To address this issue we created an in vitro analog of a feed-forward network composed of two small populations (also referred to as assemblies or layers) of living dissociated rat cortical neurons. The populations were separated by, and communicated through, a microelectromechanical systems (MEMS) device containing a strip of microscale tunnels. Delayed culturing of one population in the first layer followed by the second a few days later induced the unidirectional growth of axons through the microtunnels resulting in a primarily feed-forward communication between these two small neural populations. In this study we systematically manipulated the number of tunnels that connected each layer and hence, the number of axons providing communication between those populations. We then assess the effect of reducing the number of tunnels has upon the properties of between-layer communication capacity and fidelity of neural transmission among spike trains transmitted across and within layers. We show evidence based on Victor-Purpura's and van Rossum's spike train similarity metrics supporting the presence of both rate and temporal information embedded within these transmissions whose fidelity increased during communication both between and within layers when the number of tunnels are increased. We also provide evidence reinforcing the role of synchronized activity upon transmission fidelity during the spontaneous synchronized network burst events that propagated between layers and highlight the potential applications of these MEMs devices as a tool for further investigation of structure and functional dynamics among neural populations.

Network Events on Multiple Space and Time Scales in Cultured Neural Networks and in a Stochastic Rate Model

Cortical networks, in-vitro as well as in-vivo, can spontaneously generate a variety of collective dynamical events such as network spikes, UP and DOWN states, global oscillations, and avalanches. Though each of them has been variously recognized in previous works as expression of the excitability of the cortical tissue and the associated nonlinear dynamics, a unified picture of the determinant factors (dynamical and architectural) is desirable and not yet available. Progress has also been partially hindered by the use of a variety of statistical measures to define the network events of interest. We propose here a common probabilistic definition of network events that, applied to the firing activity of cultured neural networks, highlights the co-occurrence of network spikes, power-law distributed avalanches, and exponentially distributed 'quasi-orbits', which offer a third type of collective behavior. A rate model, including synaptic excitation and inhibition with no imposed topology, synaptic short-term depression, and finite-size noise, accounts for all these different, coexisting phenomena. We find that their emergence is largely regulated by the proximity to an oscillatory instability of the dynamics, where the non-linear excitable behavior leads to a self-amplification of activity fluctuations over a wide range of scales in space and time. In this sense, the cultured network dynamics is compatible with an excitation-inhibition balance corresponding to a slightly sub-critical regime. Finally, we propose and test a method to infer the characteristic time of the fatigue process, from the observed time course of the network's firing rate. Unlike the model, possessing a single fatigue mechanism, the cultured network appears to show multiple time scales, signalling the possible coexistence of different fatigue mechanisms.

Sloppiness in spontaneously active neuronal networks

Various plasticity mechanisms, including experience-dependent, spontaneous, as well as homeostatic ones, continuously remodel neural circuits. Yet, despite fluctuations in the properties of single neurons and synapses, the behavior and function of neuronal assemblies are generally found to be very stable over time. This raises the important question of how plasticity is coordinated across the network. To address this, we investigated the stability of network activity in cultured rat hippocampal neurons recorded with high-density multielectrode arrays over several days. We used parametric models to characterize multineuron activity patterns and analyzed their sensitivity to changes. We found that the models exhibited sloppiness, a property where the model behavior is insensitive to changes in many parameter combinations, but very sensitive to a few. The activity of neurons with sloppy parameters showed faster and larger fluctuations than the activity of a small subset of neurons associated with sensitive parameters. Furthermore, parameter sensitivity was highly correlated with firing rates. Finally, we tested our observations from cell cultures on an in vivo recording from monkey visual cortex and we confirm that spontaneous cortical activity also shows hallmarks of sloppy behavior and firing rate dependence. Our findings suggest that a small subnetwork of highly active and stable neurons supports group stability, and that this endows neuronal networks with the flexibility to continuously remodel without compromising stability and function.

Quantification of bursting and synchrony in cultured hippocampal neurons

It is widely appreciated that neuronal networks exhibit patterns of bursting and synchrony that are not captured by simple measures such as average spike rate. These patterns can encode information or represent pathological behavior such as seizures. However, methods for quantifying bursting and synchrony are not agreed upon and can be confounded with spike rate measures. Previous validation has largely relied on in silico networks and single experimental conditions. How published measures of bursting and synchrony perform when applied to biological networks of varied average spike rate and subjected to varied experimental challenges is unclear. In multielectrode array recordings of network activity, we found that two mechanistically distinct drugs, cyclothiazide and bicuculline, produced equivalent increases in average spike rate but differed in bursting and synchrony. We applied several measures of bursting to the recordings (2 threshold interval methods and a surprise-based method) and found that a measure based on an average critical interval, adjusted for the array-wide spike rate, performed best in quantifying differential drug effects. To quantify synchrony, we compared a coefficient of variation-based measure, the recently proposed spike time tiling coefficient, the SPIKE-distance measure, and a global synchrony index. The spike time tiling coefficient, the SPIKE-distance measure, and the global synchrony index all captured a difference between drugs with the best performance exhibited by the global synchrony index. In summary, our exploration should aid other investigators by highlighting strengths and limitations of current methods.

Closed-loop neuro-robotic experiments to test computational properties of neuronal networks

Information coding in the Central Nervous System (CNS) remains unexplored. There is mounting evidence that, even at a very low level, the representation of a given stimulus might be dependent on context and history. If this is actually the case, bi-directional interactions between the brain (or if need be a reduced model of it) and sensory-motor system can shed a light on how encoding and decoding of information is performed. Here an experimental system is introduced and described in which the activity of a neuronal element (i.e., a network of neurons extracted from embryonic mammalian hippocampi) is given context and used to control the movement of an artificial agent, while environmental information is fed back to the culture as a sequence of electrical stimuli. This architecture allows a quick selection of diverse encoding, decoding, and learning algorithms to test different hypotheses on the computational properties of neuronal networks.

Ontogeny of biochemical, morphological and functional parameters of synaptogenesis in primary cultures of rat hippocampal and cortical neurons

Background

Synaptogenesis is a critical neurodevelopmental process whereby pre- and postsynaptic neurons form apposed sites of contact specialized for chemical neurotransmission. Many neurodevelopmental disorders are thought to reflect altered patterns of synaptic connectivity, including imbalances between excitatory and inhibitory synapses. Developing rapid throughput approaches for assessing synaptogenesis will facilitate toxicologic and drug screening studies of neurodevelopmental disorders. The current study describes the use of high-content imaging to quantify the ontogeny of excitatory and inhibitory synapses using in vitro models of neurodevelopment. These data are compared to biochemical and functional measures of synaptogenesis.

Results

The ontogenetic patterns of synapse formation were compared between primary rodent hippocampal and cortical neurons over 28 days in vitro (DIV). As determined by ELISA, the increase in synaptophysin expression levels as cultures matured was similar between hippocampal and cortical cultures. High-content imaging of immunoreactivity of excitatory and inhibitory synaptic biomarkers demonstrated an overall greater number of synapses in hippocampal relative to cortical neurons with marked differences in the pattern of inhibitory synapse development between these two neuronal cell types. Functional assays revealed that both the mean firing rates and mean bursting rates were significantly increased in cortical cultures relative to hippocampal cultures. This difference may reflect decreased inhibitory synaptic tone in cortical versus hippocampal cultures.

Conclusions

These data demonstrate differences and similarities in the ontogeny of synaptogenesis between hippocampal and cortical neurons, depending on the biological level examined. Assessment of synaptophysin protein levels by ELISA showed a general increase in synapse formation in both cell types with increasing time in culture, while high-content imaging was able to delineate cell type-dependent differences in formation of excitatory versus inhibitory synapses. The functional significance of differences in the balance of excitatory to inhibitory synapses was confirmed by the assessment of network activity using microelectrode arrays. These results suggest that high-content imaging and microelectrode arrays provide complementary approaches for quantitative assessment of synaptogenesis, which should provide a robust readout of toxicologic and pharmacologic effects on this critical neurodevelopmental event.

Scale-free and economical features of functional connectivity in neuronal networks

A form of activity that is highly studied in cultured cortical networks is the neuronal avalanche, characterized by bursts whose distribution follows a power law. While the statistics of neuronal avalanches are well characterized, much less is known about the neuronal interactions from which they arise. We examined statistical dependencies between pairs of cells in spontaneously active cultures of cortical neurons using an information measure of transfer entropy. We show that the distribution of transfer entropy follows a power law with a slope near 3/2. Using graph-theoretic approaches of weighted networks, we demonstrate that this power law maximizes a measure of global economy that accounts for both the efficiency of neuronal interactions as well as the overall traffic in the network. Finally, we describe a pairwise Poisson model that captures the statistics of information transfer in a population of spiking neurons. Using this model, we show that avalanches can occur in systems with weak pairwise interactions, and that strong pairwise interactions can arise without avalanches, suggesting that these two measures capture distinct properties of brain dynamics.

Transfer entropy reconstruction and labeling of neuronal connections from simulated calcium imaging

Neuronal dynamics are fundamentally constrained by the underlying structural network architecture, yet much of the details of this synaptic connectivity are still unknown even in neuronal cultures in vitro. Here we extend a previous approach based on information theory, the Generalized Transfer Entropy, to the reconstruction of connectivity of simulated neuronal networks of both excitatory and inhibitory neurons. We show that, due to the model-free nature of the developed measure, both kinds of connections can be reliably inferred if the average firing rate between synchronous burst events exceeds a small minimum frequency. Furthermore, we suggest, based on systematic simulations, that even lower spontaneous inter-burst rates could be raised to meet the requirements of our reconstruction algorithm by applying a weak spatially homogeneous stimulation to the entire network. By combining multiple recordings of the same in silico network before and after pharmacologically blocking inhibitory synaptic transmission, we show then how it becomes possible to infer with high confidence the excitatory or inhibitory nature of each individual neuron.

Emergence of small-world anatomical networks in self-organizing clustered neuronal cultures

In vitro primary cultures of dissociated invertebrate neurons from locust ganglia are used to experimentally investigate the morphological evolution of assemblies of living neurons, as they self-organize from collections of separated cells into elaborated, clustered, networks. At all the different stages of the culture's development, identification of neurons' and neurites' location by means of a dedicated software allows to ultimately extract an adjacency matrix from each image of the culture. In turn, a systematic statistical analysis of a group of topological observables grants us the possibility of quantifying and tracking the progression of the main network's characteristics during the self-organization process of the culture. Our results point to the existence of a particular state corresponding to a small-world network configuration, in which several relevant graph's micro- and meso-scale properties emerge. Finally, we identify the main physical processes ruling the culture's morphological transformations, and embed them into a simplified growth model qualitatively reproducing the overall set of experimental observations.

Network bursting dynamics in excitatory cortical neuron cultures results from the combination of different adaptive mechanisms

In the brain, synchronization among cells of an assembly is a common phenomenon, and thought to be functionally relevant. Here we used an in vitro experimental model of cell assemblies, cortical cultures, combined with numerical simulations of a spiking neural network (SNN) to investigate how and why spontaneous synchronization occurs. In order to deal with excitation only, we pharmacologically blocked GABA_Aergic transmission using bicuculline. Synchronous events in cortical cultures tend to involve almost every cell and to display relatively constant durations. We have thus named these “network spikes” (NS). The inter-NS-intervals (INSIs) proved to be a more interesting phenomenon. In most cortical cultures NSs typically come in series or bursts (“bursts of NSs”, BNS), with short (∼1 s) INSIs and separated by long silent intervals (tens of s), which leads to bimodal INSI distributions. This suggests that a facilitating mechanism is at work, presumably short-term synaptic facilitation, as well as two fatigue mechanisms: one with a short timescale, presumably short-term synaptic depression, and another one with a longer timescale, presumably cellular adaptation. We thus incorporated these three mechanisms into the SNN, which, indeed, produced realistic BNSs. Next, we systematically varied the recurrent excitation for various adaptation timescales. Strong excitability led to frequent, quasi-periodic BNSs (CV∼0), and weak excitability led to rare BNSs, approaching a Poisson process (CV∼1). Experimental cultures appear to operate within an intermediate weakly-synchronized regime (CV∼0.5), with an adaptation timescale in the 2–8 s range, and well described by a Poisson-with-refractory-period model. Taken together, our results demonstrate that the INSI statistics are indeed informative: they allowed us to infer the mechanisms at work, and many parameters that we cannot access experimentally.

Selective modulation of chemical and electrical synapses of Helix neuronal networks during in vitro development

Background

A large number of invertebrate models, including the snail Helix, emerged as particularly suitable tools for investigating the formation of synapses and the specificity of neuronal connectivity. Helix neurons can be individually identified and isolated in cell culture, showing well-conserved size, position, biophysical properties, synaptic connections, and physiological functions. Although we previously showed the potential usefulness of Helix polysynaptic circuits, a full characterization of synaptic connectivity and its dynamics during network development has not been performed.

Results

In this paper, we systematically investigated the in vitro formation of polysynaptic circuits, among Helix B2 and the serotonergic C1 neurons, from a morphological and functional point of view. Since these cells are generally silent in culture, networks were chemically stimulated with either high extracellular potassium concentrations or, alternatively, serotonin. Potassium induced a transient depolarization of all neurons. On the other hand, we found prolonged firing activity, selectively maintained following the first serotonin application. Statistical analysis revealed no significant changes in neuronal dynamics during network development. Moreover, we demonstrated that the cell-selective effect of serotonin was also responsible for short-lasting alterations in C1 excitability, without long-term rebounds.

Estimation of the functional connections by means of cross-correlation analysis revealed that networks under elevated KCl concentrations exhibited strongly correlated signals with short latencies (about 5 ms), typical of electrically coupled cells. Conversely, neurons treated with serotonin were weakly connected with longer latencies (exceeding 20 ms) between the interacting neurons. Finally, we clearly demonstrated that these two types of correlations (in terms of strength/latency) were effectively related to the presence of electrical or chemical connections, by comparing Micro-Electrode Array (MEA) signal traces with intracellularly recorded cell pairs.

Conclusions

Networks treated with either potassium or serotonin were predominantly interconnected through electrical or chemical connections, respectively. Furthermore, B2 response and short-term increase in C1 excitability induced by serotonin is sufficient to trigger spontaneous activity with chemical connections, an important requisite for long-term maintenance of firing activity.

Altered network communication following a neuroprotective drug treatment

Preconditioning is defined as a range of stimuli that allow cells to withstand subsequent anaerobic and other deleterious conditions. While cell protection under preconditioning is well established, this paper investigates the influence of neuroprotective preconditioning drugs, 4-aminopyridine and bicuculline (4-AP/bic), on synaptic communication across a broad network of in vitro rat cortical neurons. Using a permutation test, we evaluated cross-correlations of extracellular spiking activity across all pairs of recording electrodes on a 64-channel multielectrode array. The resulting functional connectivity maps were analyzed in terms of their graph-theoretic properties. A small-world effect was found, characterized by a functional network with high clustering coefficient and short average path length. Twenty-four hours after exposure to 4-AP/bic, small-world properties were comparable to control cultures that were not treated with the drug. Four hours following drug washout, however, the density of functional connections increased, while path length decreased and clustering coefficient increased. These alterations in functional connectivity were maintained at four days post-washout, suggesting that 4-AP/bic preconditioning leads to long-term effects on functional networks of cortical neurons. Because of their influence on communication efficiency in neuronal networks, alterations in small-world properties hold implications for information processing in brain systems. The observed relationship between density, path length, and clustering coefficient is captured by a phenomenological model where connections are added randomly within a spatially-embedded network. Taken together, results provide information regarding functional consequences of drug therapies that are overlooked in traditional viability studies and present the first investigation of functional networks under neuroprotective preconditioning.

Emergence of a small-world functional network in cultured neurons

The functional networks of cultured neurons exhibit complex network properties similar to those found in vivo. Starting from random seeding, cultures undergo significant reorganization during the initial period in vitro, yet despite providing an ideal platform for observing developmental changes in neuronal connectivity, little is known about how a complex functional network evolves from isolated neurons. In the present study, evolution of functional connectivity was estimated from correlations of spontaneous activity. Network properties were quantified using complex measures from graph theory and used to compare cultures at different stages of development during the first 5 weeks in vitro. Networks obtained from young cultures (14 days in vitro) exhibited a random topology, which evolved to a small-world topology during maturation. The topology change was accompanied by an increased presence of highly connected areas (hubs) and network efficiency increased with age. The small-world topology balances integration of network areas with segregation of specialized processing units. The emergence of such network structure in cultured neurons, despite a lack of external input, points to complex intrinsic biological mechanisms. Moreover, the functional network of cultures at mature ages is efficient and highly suited to complex processing tasks.

Modeling the impact of common noise inputs on the network activity of retinal ganglion cells

Synchronized spontaneous firing among retinal ganglion cells (RGCs), on timescales faster than visual responses, has been reported in many studies. Two candidate mechanisms of synchronized firing include direct coupling and shared noisy inputs. In neighboring parasol cells of primate retina, which exhibit rapid synchronized firing that has been studied extensively, recent experimental work indicates that direct electrical or synaptic coupling is weak, but shared synaptic input in the absence of modulated stimuli is strong. However, previous modeling efforts have not accounted for this aspect of firing in the parasol cell population. Here we develop a new model that incorporates the effects of common noise, and apply it to analyze the light responses and synchronized firing of a large, densely-sampled network of over 250 simultaneously recorded parasol cells. We use a generalized linear model in which the spike rate in each cell is determined by the linear combination of the spatio-temporally filtered visual input, the temporally filtered prior spikes of that cell, and unobserved sources representing common noise. The model accurately captures the statistical structure of the spike trains and the encoding of the visual stimulus, without the direct coupling assumption present in previous modeling work. Finally, we examined the problem of decoding the visual stimulus from the spike train given the estimated parameters. The common-noise model produces Bayesian decoding performance as accurate as that of a model with direct coupling, but with significantly more robustness to spike timing perturbations.

NETWORK DYNAMICS AND SYNCHRONOUS ACTIVITY IN CULTURED CORTICAL NEURONS

Neurons extracted from specific areas of the Central Nervous System (CNS), such as the hippocampus, the cortex and the spinal cord, can be cultured in vitro and coupled with a micro-electrode array (MEA) for months. After a few days, neurons connect each other with functionally active synapses, forming a random network and displaying spontaneous electrophysiological activity. In spite of their simplified level of organization, they represent an useful framework to study general information processing properties and specific basic learning mechanisms in the nervous system. These experimental preparations show patterns of collective rhythmic activity characterized by burst and spike firing. The patterns of electrophysiological activity may change as a consequence of external stimulation (i.e., chemical and/or electrical inputs) and by partly modifying the "randomness" of the network architecture (i.e., confining neuronal sub-populations in clusters with micro-machined barriers). In particular we investigated how the spontaneous rhythmic and synchronous activity can be modulated or drastically changed by focal electrical stimulation, pharmacological manipulation and network segregation. Our results show that burst firing and global synchronization can be enhanced or reduced; and that the degree of synchronous activity in the network can be characterized by simple parameters such as cross-correlation on burst events.

Inferring network dynamics and neuron properties from population recordings

Understanding the computational capabilities of the nervous system means to “identify” its emergent multiscale dynamics. For this purpose, we propose a novel model-driven identification procedure and apply it to sparsely connected populations of excitatory integrate-and-fire neurons with spike frequency adaptation (SFA). Our method does not characterize the system from its microscopic elements in a bottom-up fashion, and does not resort to any linearization. We investigate networks as a whole, inferring their properties from the response dynamics of the instantaneous discharge rate to brief and aspecific supra-threshold stimulations. While several available methods assume generic expressions for the system as a black box, we adopt a mean-field theory for the evolution of the network transparently parameterized by identified elements (such as dynamic timescales), which are in turn non-trivially related to single-neuron properties. In particular, from the elicited transient responses, the input–output gain function of the neurons in the network is extracted and direct links to the microscopic level are made available: indeed, we show how to extract the decay time constant of the SFA, the absolute refractory period and the average synaptic efficacy. In addition and contrary to previous attempts, our method captures the system dynamics across bifurcations separating qualitatively different dynamical regimes. The robustness and the generality of the methodology is tested on controlled simulations, reporting a good agreement between theoretically expected and identified values. The assumptions behind the underlying theoretical framework make the method readily applicable to biological preparations like cultured neuron networks and in vitro brain slices.

Neurobiologically realistic determinants of self-organized criticality in networks of spiking neurons

Self-organized criticality refers to the spontaneous emergence of self-similar dynamics in complex systems poised between order and randomness. The presence of self-organized critical dynamics in the brain is theoretically appealing and is supported by recent neurophysiological studies. Despite this, the neurobiological determinants of these dynamics have not been previously sought. Here, we systematically examined the influence of such determinants in hierarchically modular networks of leaky integrate-and-fire neurons with spike-timing-dependent synaptic plasticity and axonal conduction delays. We characterized emergent dynamics in our networks by distributions of active neuronal ensemble modules (neuronal avalanches) and rigorously assessed these distributions for power-law scaling. We found that spike-timing-dependent synaptic plasticity enabled a rapid phase transition from random subcritical dynamics to ordered supercritical dynamics. Importantly, modular connectivity and low wiring cost broadened this transition, and enabled a regime indicative of self-organized criticality. The regime only occurred when modular connectivity, low wiring cost and synaptic plasticity were simultaneously present, and the regime was most evident when between-module connection density scaled as a power-law. The regime was robust to variations in other neurobiologically relevant parameters and favored systems with low external drive and strong internal interactions. Increases in system size and connectivity facilitated internal interactions, permitting reductions in external drive and facilitating convergence of postsynaptic-response magnitude and synaptic-plasticity learning rate parameter values towards neurobiologically realistic levels. We hence infer a novel association between self-organized critical neuronal dynamics and several neurobiologically realistic features of structural connectivity. The central role of these features in our model may reflect their importance for neuronal information processing.