Interplay between population firing stability and single neuron dynamics in hippocampal networks

Unsupervised restoration of a complex learned behavior after large-scale neuronal perturbation

Reliable execution of precise behaviors requires that brain circuits are resilient to variations in neuronal dynamics. Genetic perturbation of the majority of excitatory neurons in HVC, a brain region involved in song production, in adult songbirds with stereotypical songs triggered severe degradation of the song. The song fully recovered within 2 weeks, and substantial improvement occurred even when animals were prevented from singing during the recovery period, indicating that offline mechanisms enable recovery in an unsupervised manner. Song restoration was accompanied by increased excitatory synaptic input to neighboring, unmanipulated neurons in the same brain region. A model inspired by the behavioral and electrophysiological findings suggests that unsupervised single-cell and population-level homeostatic plasticity rules can support the functional restoration after large-scale disruption of networks that implement sequential dynamics. These observations suggest the existence of cellular and systems-level restorative mechanisms that ensure behavioral resilience.

Cortical reactivations predict future sensory responses

Many theories of offline memory consolidation posit that the pattern of neurons activated during a salient sensory experience will be faithfully reactivated, thereby stabilizing the pattern1,2. However, sensory-evoked patterns are not stable but, instead, drift across repeated experiences3-6. Here, to investigate the relationship between reactivations and the drift of sensory representations, we imaged the calcium activity of thousands of excitatory neurons in the mouse lateral visual cortex. During the minute after a visual stimulus, we observed transient, stimulus-specific reactivations, often coupled with hippocampal sharp-wave ripples. Stimulus-specific reactivations were abolished by local cortical silencing during the preceding stimulus. Reactivations early in a session systematically differed from the pattern evoked by the previous stimulus-they were more similar to future stimulus response patterns, thereby predicting both within-day and across-day representational drift. In particular, neurons that participated proportionally more or less in early stimulus reactivations than in stimulus response patterns gradually increased or decreased their future stimulus responses, respectively. Indeed, we could accurately predict future changes in stimulus responses and the separation of responses to distinct stimuli using only the rate and content of reactivations. Thus, reactivations may contribute to a gradual drift and separation in sensory cortical response patterns, thereby enhancing sensory discrimination7.

Early alterations in the MCH system link aberrant neuronal activity and sleep disturbances in a mouse model of Alzheimer's disease

Early Alzheimer's disease (AD) is associated with hippocampal hyperactivity and decreased sleep quality. Here we show that homeostatic mechanisms transiently counteract the increased excitatory drive to CA1 neurons in App^NL-G-F mice, but that this mechanism fails in older mice. Spatial transcriptomics analysis identifies Pmch as part of the adaptive response in App^NL-G-F mice. Pmch encodes melanin-concentrating hormone (MCH), which is produced in sleep-active lateral hypothalamic neurons that project to CA1 and modulate memory. We show that MCH downregulates synaptic transmission, modulates firing rate homeostasis in hippocampal neurons and reverses the increased excitatory drive to CA1 neurons in App^NL-G-F mice. App^NL-G-F mice spend less time in rapid eye movement (REM) sleep. App^NL-G-F mice and individuals with AD show progressive changes in morphology of CA1-projecting MCH axons. Our findings identify the MCH system as vulnerable in early AD and suggest that impaired MCH-system function contributes to aberrant excitatory drive and sleep defects, which can compromise hippocampus-dependent functions.

Large-Area Field Potential Imaging Having Single Neuron Resolution Using 236 880 Electrodes CMOS-MEA Technology

The electrophysiological technology having a high spatiotemporal resolution at the single-cell level and noninvasive measurements of large areas provide insights on underlying neuronal function. Here, a complementary metal-oxide semiconductor (CMOS)-microelectrode array (MEA) is used that uses 236 880 electrodes each with an electrode size of 11.22 × 11.22 µm and 236 880 covering a wide area of 5.5 × 5.9 mm in presenting a detailed and single-cell-level neural activity analysis platform for brain slices, human iPS cell-derived cortical networks, peripheral neurons, and human brain organoids. Propagation pattern characteristics between brain regions changes the synaptic propagation into compounds based on single-cell time-series patterns, classification based on single DRG neuron firing patterns and compound responses, axonal conduction characteristics and changes to anticancer drugs, and network activities and transition to compounds in brain organoids are extracted. This detailed analysis of neural activity at the single-cell level using the CMOS-MEA provides a new understanding of the basic mechanisms of brain circuits in vitro and ex vivo, on human neurological diseases for drug discovery, and compound toxicity assessment.

Creation of Neuronal Ensembles and Cell-Specific Homeostatic Plasticity through Chronic Sparse Optogenetic Stimulation

Cortical computations emerge from the dynamics of neurons embedded in complex cortical circuits. Within these circuits, neuronal ensembles, which represent subnetworks with shared functional connectivity, emerge in an experience-dependent manner. Here we induced ensembles in ex vivo cortical circuits from mice of either sex by differentially activating subpopulations through chronic optogenetic stimulation. We observed a decrease in voltage correlation, and importantly a synaptic decoupling between the stimulated and nonstimulated populations. We also observed a decrease in firing rate during Up-states in the stimulated population. These ensemble-specific changes were accompanied by decreases in intrinsic excitability in the stimulated population, and a decrease in connectivity between stimulated and nonstimulated pyramidal neurons. By incorporating the empirically observed changes in intrinsic excitability and connectivity into a spiking neural network model, we were able to demonstrate that changes in both intrinsic excitability and connectivity accounted for the decreased firing rate, but only changes in connectivity accounted for the observed decorrelation. Our findings help ascertain the mechanisms underlying the ability of chronic patterned stimulation to create ensembles within cortical circuits and, importantly, show that while Up-states are a global network-wide phenomenon, functionally distinct ensembles can preserve their identity during Up-states through differential firing rates and correlations.SIGNIFICANCE STATEMENT The connectivity and activity patterns of local cortical circuits are shaped by experience. This experience-dependent reorganization of cortical circuits is driven by complex interactions between different local learning rules, external input, and reciprocal feedback between many distinct brain areas. Here we used an ex vivo approach to demonstrate how simple forms of chronic external stimulation can shape local cortical circuits in terms of their correlated activity and functional connectivity. The absence of feedback between different brain areas and full control of external input allowed for a tractable system to study the underlying mechanisms and development of a computational model. Results show that differential stimulation of subpopulations of neurons significantly reshapes cortical circuits and forms subnetworks referred to as neuronal ensembles.

Paradoxical self-sustained dynamics emerge from orchestrated excitatory and inhibitory homeostatic plasticity rules

Cortical networks have the remarkable ability to self-assemble into dynamic regimes in which excitatory positive feedback is balanced by recurrent inhibition. This inhibition-stabilized regime is increasingly viewed as the default dynamic regime of the cortex, but how it emerges in an unsupervised manner remains unknown. We prove that classic forms of homeostatic plasticity are unable to drive recurrent networks to an inhibition-stabilized regime due to the well-known paradoxical effect. We next derive a novel family of cross-homeostatic rules that lead to the unsupervised emergence of inhibition-stabilized networks. These rules shed new light on how the brain may reach its default dynamic state and provide a valuable tool to self-assemble artificial neural networks into ideal computational regimes.

Self-sustained neural activity maintained through local recurrent connections is of fundamental importance to cortical function. Converging theoretical and experimental evidence indicates that cortical circuits generating self-sustained dynamics operate in an inhibition-stabilized regime. Theoretical work has established that four sets of weights (W_E←E, W_E←I, W_I←E, and W_I←I) must obey specific relationships to produce inhibition-stabilized dynamics, but it is not known how the brain can appropriately set the values of all four weight classes in an unsupervised manner to be in the inhibition-stabilized regime. We prove that standard homeostatic plasticity rules are generally unable to generate inhibition-stabilized dynamics and that their instability is caused by a signature property of inhibition-stabilized networks: the paradoxical effect. In contrast, we show that a family of “cross-homeostatic” rules overcome the paradoxical effect and robustly lead to the emergence of stable dynamics. This work provides a model of how—beginning from a silent network—self-sustained inhibition-stabilized dynamics can emerge from learning rules governing all four synaptic weight classes in an orchestrated manner.

Excitatory and inhibitory hippocampal neurons differ in their homeostatic adaptation to chronic M-channel modulation

A large body of studies has investigated bidirectional homeostatic plasticity both in vitro and in vivo using numerous pharmacological manipulations of activity or behavioral paradigms. However, these experiments rarely explored in the same cellular system the bidirectionality of the plasticity and simultaneously on excitatory and inhibitory neurons. M-channels are voltage-gated potassium channels that play a crucial role in regulating neuronal excitability and plasticity. In cultured hippocampal excitatory neurons, we previously showed that chronic exposure to the M-channel blocker XE991 leads to adaptative compensations, thereby triggering at different timescales intrinsic and synaptic homeostatic plasticity. This plastic adaptation barely occurs in hippocampal inhibitory neurons. In this study, we examined whether this homeostatic plasticity induced by M-channel inhibition was bidirectional by investigating the acute and chronic effects of the M-channel opener retigabine on hippocampal neuronal excitability. Acute retigabine exposure decreased excitability in both excitatory and inhibitory neurons. Chronic retigabine treatment triggered in excitatory neurons homeostatic adaptation of the threshold current and spontaneous firing rate at a time scale of 4–24 h. These plastic changes were accompanied by a substantial decrease in the M-current density and by a small, though significant, proximal relocation of Kv7.3-FGF14 segment along the axon initial segment. Thus, bidirectional homeostatic changes were observed in excitatory neurons though not symmetric in kinetics and mechanisms. Contrastingly, in inhibitory neurons, the compensatory changes in intrinsic excitability barely occurred after 48 h, while no homeostatic normalization of the spontaneous firing rate was observed. Our results indicate that excitatory and inhibitory hippocampal neurons differ in their adaptation to chronic alterations in neuronal excitability induced by M-channel bidirectional modulation.

Altered stability of brain functional architecture after sleep deprivation: A resting-state functional magnetic resonance imaging study

Background and aims

Resting-state functional magnetic resonance imaging (fMRI) studies using static and dynamic functional connectivity (FC) approaches have revealed brain dysfunction resulting from sleep deprivation (SD). The effects of SD on the stability of brain functional architecture remain unclear. This study investigated the functional stability (FS) changes induced by SD and its association with neurocognitive alterations.

Materials and methods

In this study, we recruited 24 healthy women. All participants underwent two sessions of resting-state fMRI scanning and neurocognitive assessment. The assessments included the Digit Symbol Test, Digit Span Test, Trail-Making Test (TMT), and Complex Figure Test (CFT). Participants completed one session under rested wakefulness (RW) and one session after SD for 24 h. To estimate dynamic FC, we used the sliding window approach; and then, to characterize the FS of each voxel, we measured dynamic FC concordance over time. We used a paired t-test to identify differences in FS between RW and SD. To examine the relationship between these changes in FS and alterations in neurocognitive performance, we conducted Spearman’s correlation analyses.

Results

SD affected the performance of the Digit Symbol Test, Digit Span Test, and CFT. Compared with RW, subjects with SD exhibited decreased FS in the bilateral anterior and posterior cingulate gyrus and medial frontal gyrus, right superior frontal gyrus, and cerebellum posterior lobe, while they exhibited increased FS in the bilateral precentral/postcentral gyrus and supplementary motor area, right parahippocampal gyrus and fusiform gyrus, left inferior occipital gyrus, and bilateral cerebellum anterior lobe. After SD, FS changes in the right parahippocampal gyrus and fusiform gyrus were correlated with altered performance in the Digit Symbol Test and CFT.

Conclusion

Our findings showed that the stability of the brain’s functional architecture could be altered by SD. This stability alteration may correspond to multiple neurocognitive domain changes.

IGF-1 receptor regulates upward firing rate homeostasis via the mitochondrial calcium uniporter

An emerging hypothesis is that neuronal circuits homeostatically maintain a stable spike rate despite continuous environmental changes. This firing rate homeostasis is believed to confer resilience to neurodegeneration and cognitive decline. We show that insulin-like growth factor-1 receptor (IGF-1R) is necessary for homeostatic response of mean firing rate to inactivity, termed “upward firing rate homeostasis.” We show that its mechanism of action is to couple spike bursts with downstream mitochondrial Ca²⁺ influx via the mitochondrial calcium uniporter complex (MCUc). We propose that MCUc is a homeostatic Ca²⁺ sensor that triggers the integrated homeostatic response. Firing rate homeostasis may be the principal mechanism by which IGF-1R regulates aging and neurodevelopmental and neurodegenerative disorders.

Regulation of firing rate homeostasis constitutes a fundamental property of central neural circuits. While intracellular Ca²⁺ has long been hypothesized to be a feedback control signal, the molecular machinery enabling a network-wide homeostatic response remains largely unknown. We show that deletion of insulin-like growth factor-1 receptor (IGF-1R) limits firing rate homeostasis in response to inactivity, without altering the distribution of baseline firing rates. The deficient firing rate homeostatic response was due to disruption of both postsynaptic and intrinsic plasticity. At the cellular level, we detected a fraction of IGF-1Rs in mitochondria, colocalized with the mitochondrial calcium uniporter complex (MCUc). IGF-1R deletion suppressed transcription of the MCUc members and burst-evoked mitochondrial Ca²⁺ (mitoCa²⁺) by weakening mitochondria-to-cytosol Ca²⁺ coupling. Overexpression of either mitochondria-targeted IGF-1R or MCUc in IGF-1R–deficient neurons was sufficient to rescue the deficits in burst-to-mitoCa²⁺ coupling and firing rate homeostasis. Our findings indicate that mitochondrial IGF-1R is a key regulator of the integrated homeostatic response by tuning the reliability of burst transfer by MCUc. Based on these results, we propose that MCUc acts as a homeostatic Ca²⁺ sensor. Faulty activation of MCUc may drive dysregulation of firing rate homeostasis in aging and in brain disorders associated with aberrant IGF-1R/MCUc signaling.

Aβ/Amyloid Precursor Protein-Induced Hyperexcitability and Dysregulation of Homeostatic Synaptic Plasticity in Neuron Models of Alzheimer’s Disease

Alzheimer’s disease (AD) is increasingly seen as a disease of synapses and diverse evidence has implicated the amyloid-β peptide (Aβ) in synapse damage. The molecular and cellular mechanism(s) by which Aβ and/or its precursor protein, the amyloid precursor protein (APP) can affect synapses remains unclear. Interestingly, early hyperexcitability has been described in human AD and mouse models of AD, which precedes later hypoactivity. Here we show that neurons in culture with either elevated levels of Aβ or with human APP mutated to prevent Aβ generation can both induce hyperactivity as detected by elevated calcium transient frequency and amplitude. Since homeostatic synaptic plasticity (HSP) mechanisms normally maintain a setpoint of activity, we examined whether HSP was altered in AD transgenic neurons. Using methods known to induce HSP, we demonstrate that APP protein levels are regulated by chronic modulation of activity and that AD transgenic neurons have an impaired adaptation of calcium transients to global changes in activity. Further, AD transgenic compared to WT neurons failed to adjust the length of their axon initial segments (AIS), an adaptation known to alter excitability. Thus, we show that both APP and Aβ influence neuronal activity and that mechanisms of HSP are disrupted in primary neuron models of AD.

Coupling of in vitro Neocortical-Hippocampal Coculture Bursts Induces Different Spike Rhythms in Individual Networks

Brain-state alternation is important for long-term memory formation. Each brain state can be identified with a specific process in memory formation, e.g., encoding during wakefulness or consolidation during sleeping. The hippocampal-neocortical dialogue was proposed as a hypothetical framework for systems consolidation, which features different cross-frequency couplings between the hippocampus and distributed neocortical regions in different brain states. Despite evidence supporting this hypothesis, little has been reported about how information is processed with shifts in brain states. To address this gap, we developed an in vitro neocortical-hippocampal coculture model to study how activity coupling can affect connections between coupled networks. Neocortical and hippocampal neurons were cultured in two different compartments connected by a micro-tunnel structure. The network activity of the coculture model was recorded by microelectrode arrays underlying the substrate. Rhythmic bursting was observed in the spontaneous activity and electrical evoked responses. Rhythmic bursting activity in one compartment could couple to that in the other via axons passing through the micro-tunnels. Two types of coupling patterns were observed: slow-burst coupling (neocortex at 0.1–0.5 Hz and hippocampus at 1 Hz) and fast burst coupling (neocortex at 20–40 Hz and hippocampus at 4–10 Hz). The network activity showed greater synchronicity in the slow-burst coupling, as indicated by changes in the burstiness index. Network synchronicity analysis suggests the presence of different information processing states under different burst activity coupling patterns. Our results suggest that the hippocampal-neocortical coculture model possesses multiple modes of burst activity coupling between the cortical and hippocampal parts. With the addition of external stimulation, the neocortical-hippocampal network model we developed can elucidate the influence of state shifts on information processing.

A feedback control principle common to several biological and engineered systems

Feedback control is used by many distributed systems to optimize behaviour. Traditional feedback control algorithms spend significant resources to constantly sense and stabilize a continuous control variable of interest, such as vehicle speed for implementing cruise control, or body temperature for maintaining homeostasis. By contrast, discrete-event feedback (e.g. a server acknowledging when data are successfully transmitted, or a brief antennal interaction when an ant returns to the nest after successful foraging) can reduce costs associated with monitoring a continuous variable; however, optimizing behaviour in this setting requires alternative strategies. Here, we studied parallels between discrete-event feedback control strategies in biological and engineered systems. We found that two common engineering rules—additive-increase, upon positive feedback, and multiplicative-decrease, upon negative feedback, and multiplicative-increase multiplicative-decrease—are used by diverse biological systems, including for regulating foraging by harvester ant colonies, for maintaining cell-size homeostasis, and for synaptic learning and adaptation in neural circuits. These rules support several goals of these systems, including optimizing efficiency (i.e. using all available resources); splitting resources fairly among cooperating agents, or conversely, acquiring resources quickly among competing agents; and minimizing the latency of responses, especially when conditions change. We hypothesize that theoretical frameworks from distributed computing may offer new ways to analyse adaptation behaviour of biology systems, and in return, biological strategies may inspire new algorithms for discrete-event feedback control in engineering.

The impact of closed-loop intracortical stimulation on neural activity in brain-injured, anesthetized animals

BACKGROUND

Acquired brain injuries, such as stroke, are a major cause of long-term disability worldwide. Intracortical microstimulation (ICMS) can be used successfully to assist in guiding appropriate connections to restore lost sensorimotor integration. Activity-Dependent Stimulation (ADS) is a specific type of closed-loop ICMS that aims at coupling the activity of two different brain regions by stimulating one in response to activity in the other. Recently, ADS was used to effectively promote behavioral recovery in rodent models following a unilateral traumatic brain injury in the primary motor cortex. While behavioral benefits have been described, the neurophysiological changes in spared areas in response to this type of stimulation have not been fully characterized. Here we explored how single-unit spiking activity is impacted by a focal ischemic lesion and, subsequently, by an ADS treatment.

METHODS

Intracortical microelectrode arrays were implanted in the ipsilesional rostral forelimb area (RFA) to record spike activity and to trigger intracortical microstimulation in the primary somatosensory area (S1) of anaesthetized Long Evans rats. An ischemic injury was induced in the caudal forelimb area through microinjections of Endothelin-1. Activity from both RFA and S1 was recorded and analyzed off-line by evaluating possible changes, either induced by the lesion in the Control group or by stimulation in the ADS group.

RESULTS

We found that the ischemic lesion in the motor area led to an overall increase in spike activity within RFA and a decrease in S1 with respect to the baseline condition. Subsequent treatment with ADS increased the firing rate in both RFA and S1. Post-stimulation spiking activity was significantly higher compared to pre-stimulation activity in the ADS animals versus non-stimulated controls. Moreover, stimulation promoted the generation of highly synchronized bursting patterns in both RFA and S1 only in the ADS group.

CONCLUSIONS

This study describes the impact on single-unit activity in ipsilesional areas immediately following a cortical infarct and demonstrates that application of ADS is effective in altering this activity.

Visual association cortex links cues with conjunctions of reward and locomotor contexts

Postrhinal cortex (POR) and neighboring lateral visual association areas are necessary for identifying objects and interpreting them in specific contexts, but how POR neurons encode the same object across contexts remains unclear. Here, we imaged excitatory neurons in mouse POR across tens of days prior to and throughout initial cue-reward learning and reversal learning. We assessed responses to the same cue when it was rewarded or unrewarded, during both locomotor and stationary contexts. Surprisingly, a large class of POR neurons were minimally cue-driven prior to learning. After learning, distinct clusters within this class responded selectively to a given cue when presented in a specific conjunction of reward and locomotion contexts. In addition, another class contained clusters of neurons whose cue responses were more transient, insensitive to reward learning, and adapted over thousands of presentations. These two classes of POR neurons may support context-dependent interpretation and context-independent identification of sensory cues.

Self-healing codes: How stable neural populations can track continually reconfiguring neural representations

The brain is capable of adapting while maintaining stable long-term memories and learned skills. Recent experiments show that neural responses are highly plastic in some circuits, while other circuits maintain consistent responses over time, raising the question of how these circuits interact coherently. We show how simple, biologically motivated Hebbian and homeostatic mechanisms in single neurons can allow circuits with fixed responses to continuously track a plastic, changing representation without reference to an external learning signal.

As an adaptive system, the brain must retain a faithful representation of the world while continuously integrating new information. Recent experiments have measured population activity in cortical and hippocampal circuits over many days and found that patterns of neural activity associated with fixed behavioral variables and percepts change dramatically over time. Such “representational drift” raises the question of how malleable population codes can interact coherently with stable long-term representations that are found in other circuits and with relatively rigid topographic mappings of peripheral sensory and motor signals. We explore how known plasticity mechanisms can allow single neurons to reliably read out an evolving population code without external error feedback. We find that interactions between Hebbian learning and single-cell homeostasis can exploit redundancy in a distributed population code to compensate for gradual changes in tuning. Recurrent feedback of partially stabilized readouts could allow a pool of readout cells to further correct inconsistencies introduced by representational drift. This shows how relatively simple, known mechanisms can stabilize neural tuning in the short term and provides a plausible explanation for how plastic neural codes remain integrated with consolidated, long-term representations.

Disrupted neural correlates of anesthesia and sleep reveal early circuit dysfunctions in Alzheimer models

Dysregulated homeostasis of neural activity has been hypothesized to drive Alzheimer's disease (AD) pathogenesis. AD begins with a decades-long presymptomatic phase, but whether homeostatic mechanisms already begin failing during this silent phase is unknown. We show that before the onset of memory decline and sleep disturbances, familial AD (fAD) model mice display no deficits in CA1 mean firing rate (MFR) during active wakefulness. However, homeostatic down-regulation of CA1 MFR is disrupted during non-rapid eye movement (NREM) sleep and general anesthesia in fAD mouse models. The resultant hyperexcitability is attenuated by the mitochondrial dihydroorotate dehydrogenase (DHODH) enzyme inhibitor, which tunes MFR toward lower set-point values. Ex vivo fAD mutations impair downward MFR homeostasis, resulting in pathological MFR set points in response to anesthetic drug and inhibition blockade. Thus, firing rate dyshomeostasis of hippocampal circuits is masked during active wakefulness but surfaces during low-arousal brain states, representing an early failure of the silent disease stage.

REST/NRSF drives homeostatic plasticity of inhibitory synapses in a target-dependent fashion

The repressor-element 1-silencing transcription/neuron-restrictive silencer factor (REST/NRSF) controls hundreds of neuron-specific genes. We showed that REST/NRSF downregulates glutamatergic transmission in response to hyperactivity, thus contributing to neuronal homeostasis. However, whether GABAergic transmission is also implicated in the homeostatic action of REST/NRSF is unknown. Here, we show that hyperactivity-induced REST/NRSF activation, triggers a homeostatic rearrangement of GABAergic inhibition, with increased frequency of miniature inhibitory postsynaptic currents (IPSCs) and amplitude of evoked IPSCs in mouse cultured hippocampal neurons. Notably, this effect is limited to inhibitory-onto-excitatory neuron synapses, whose density increases at somatic level and decreases in dendritic regions, demonstrating a complex target- and area-selectivity. The upscaling of perisomatic inhibition was occluded by TrkB receptor inhibition and resulted from a coordinated and sequential activation of the Npas4 and Bdnf gene programs. On the opposite, the downscaling of dendritic inhibition was REST-dependent, but BDNF-independent. The findings highlight the central role of REST/NRSF in the complex transcriptional responses aimed at rescuing physiological levels of network activity in front of the ever-changing environment.

A Correspondence Between Normalization Strategies in Artificial and Biological Neural Networks

A fundamental challenge at the interface of machine learning and neuroscience is to uncover computational principles that are shared between artificial and biological neural networks. In deep learning, normalization methods such as batch normalization, weight normalization, and their many variants help to stabilize hidden unit activity and accelerate network training, and these methods have been called one of the most important recent innovations for optimizing deep networks. In the brain, homeostatic plasticity represents a set of mechanisms that also stabilize and normalize network activity to lie within certain ranges, and these mechanisms are critical for maintaining normal brain function. In this article, we discuss parallels between artificial and biological normalization methods at four spatial scales: normalization of a single neuron's activity, normalization of synaptic weights of a neuron, normalization of a layer of neurons, and normalization of a network of neurons. We argue that both types of methods are functionally equivalent-that is, both push activation patterns of hidden units toward a homeostatic state, where all neurons are equally used-and we argue that such representations can improve coding capacity, discrimination, and regularization. As a proof of concept, we develop an algorithm, inspired by a neural normalization technique called synaptic scaling, and show that this algorithm performs competitively against existing normalization methods on several data sets. Overall, we hope this bidirectional connection will inspire neuroscientists and machine learners in three ways: to uncover new normalization algorithms based on established neurobiological principles; to help quantify the trade-offs of different homeostatic plasticity mechanisms used in the brain; and to offer insights about how stability may not hinder, but may actually promote, plasticity.

Representational drift in the mouse visual cortex

Recent studies have shown that neuronal representations gradually change over time despite no changes in the stimulus, environment, or behavior. However, such representational drift has been assumed to be a property of high-level brain structures, whereas earlier circuits, such as sensory cortices, have been assumed to stably encode information over time. Here, we analyzed large-scale optical and electrophysiological recordings from six visual cortical areas in behaving mice that were repeatedly presented with the same natural movies. Contrary to the prevailing notion, we found representational drift over timescales spanning minutes to days across multiple visual areas, cortical layers, and cell types. Notably, neural-code stability did not reflect the hierarchy of information flow across areas. Although individual neurons showed time-dependent changes in their coding properties, the structure of the relationships between population activity patterns remained stable and stereotypic. Such population-level organization may underlie stable visual perception despite continuous changes in neuronal responses.

Network-centered homeostasis through inhibition maintains hippocampal spatial map and cortical circuit function

Despite ongoing experiential change, neural activity maintains remarkable stability. Although this is thought to be mediated by homeostatic plasticity, what aspect of neural activity is conserved and how the flexibility necessary for learning and memory is maintained is not fully understood. Experimental studies suggest that there exists network-centered, in addition to the well-studied neuron-centered, control. Here we computationally study such a potential mechanism: input-dependent inhibitory plasticity (IDIP). In a hippocampal model, we show that IDIP can explain the emergence of active and silent place cells as well as remapping following silencing of active place cells. Furthermore, we show that IDIP can also stabilize recurrent dynamics while preserving firing rate heterogeneity and stimulus representation, as well as persistent activity after memory encoding. Hence, the establishment of global network balance with IDIP has diverse functional implications and may be able to explain experimental phenomena across different brain areas.

•

Input-dependent inhibitory plasticity (IDIP) provides network-wide homeostasis

•

IDIP can explain hippocampal remapping following place map silencing

•

IDIP can also provide recurrent network homeostasis with firing rate diversity

Towards proteinoid computers. Hypothesis paper

Proteinoids - thermal proteins - are produced by heating amino acids to their melting point and initiation of polymerisation to produce polymeric chains. Proteinoids swell in aqueous solution into hollow microspheres. The proteinoid microspheres produce endogenous burst of electrical potential spikes and change patterns of their electrical activity in response to illumination. The microspheres can interconnect by pores and tubes and form networks with a programmable growth. We speculate on how ensembles of the proteinoid microspheres can be developed into unconventional computing devices.

CaMKIV Signaling Is Not Essential for the Maintenance of Intrinsic or Synaptic Properties in Mouse Visual Cortex

Pyramidal neurons in rodent visual cortex homeostatically maintain their firing rates in vivo within a target range. In young cultured rat cortical neurons, Ca²⁺/calmodulin-dependent kinase IV (CaMKIV) signaling jointly regulates excitatory synaptic strength and intrinsic excitability to allow neurons to maintain their target firing rate. However, the role of CaMKIV signaling in regulating synaptic strength and intrinsic excitability in vivo has not been tested. Here, we show that in pyramidal neurons in visual cortex of juvenile male and female mice, CaMKIV signaling is not essential for the maintenance of basal synaptic or intrinsic properties. Neither CaMKIV conditional knock-down nor viral expression of dominant negative CaMKIV (dnCaMKIV) in vivo disrupts the intrinsic excitability or synaptic input strength of pyramidal neurons in primary visual cortex (V1), and CaMKIV signaling is not required for the increase in intrinsic excitability seen following monocular deprivation (MD). Viral expression of constitutively active CaMKIV (caCaMKIV) in vivo causes a complex disruption of the neuronal input/output function but does not affect synaptic input strength. Taken together, these results demonstrate that although augmented in vivo CaMKIV signaling can alter neuronal excitability, either endogenous CaMKIV signaling is dispensable for maintenance of excitability, or impaired CaMKIV signaling is robustly compensated.

Riluzole ameliorates soluble Aβ1-42-induced impairments in spatial memory by modulating the glutamatergic/GABAergic balance in the dentate gyrus

Soluble amyloid beta (Aβ) is believed to contribute to cognitive deficits in the early stages of Alzheimer's disease (AD). Increased soluble Aβ_1-42 in the hippocampus is closely correlated with spatial learning and memory deficits in AD. Riluzole (RLZ), an FDA-approved drug for amyotrophic lateral sclerosis (ALS), has beneficial effects for AD. However, the mechanism underlying the effects remains unclear. In this study, its neuroprotective effect against soluble Aβ_1-42-induced spatial cognitive deficits in rats was assessed. We found that intrahippocampal injection of soluble Aβ_1-42 impaired spatial cognitive function and suppressed long-term potentiation (LTP) of the DG region, which was relevant to soluble Aβ_1-42-induced shift of the hippocampal excitation/inhibition balance toward excitation. Interestingly, RLZ ameliorated Aβ_1-42-induced behavioral and LTP impairments through rescuing the soluble Aβ_1-42-induced excitation/inhibition imbalance. RLZ attenuated Aβ_1-42-mediated facilitation of excitatory synaptic transmission by facilitating glutamate reuptake and decreasing presynaptic glutamate release. Meanwhile, RLZ attenuated the suppression of inhibitory synaptic transmission caused by Aβ_1-42 by potentiating postsynaptic GABA receptor function. These results suggest that RLZ exerts a neuroprotective effect against soluble Aβ_1-42-related spatial cognitive deficits through rescuing the excitation/inhibition imbalance, and it could be a potential therapy for AD.

Restoring neuronal chloride homeostasis with anti-NKCC1 gene therapy rescues cognitive deficits in a mouse model of Down syndrome

A common feature of diverse brain disorders is the alteration of GABA-mediated inhibition because of aberrant, intracellular chloride homeostasis induced by changes in the expression and/or function of chloride transporters. Notably, pharmacological inhibition of the chloride importer NKCC1 is able to rescue brain-related core deficits in animal models of these pathologies and in some human clinical studies. Here, we show that reducing NKCC1 expression by RNA interference in the Ts65Dn mouse model of Down syndrome (DS) restores intracellular chloride concentration, efficacy of gamma-aminobutyric acid (GABA)-mediated inhibition, and neuronal network dynamics in vitro and ex vivo. Importantly, adeno-associated virus (AAV)-mediated, neuron-specific NKCC1 knockdown in vivo rescues cognitive deficits in diverse behavioral tasks in Ts65Dn animals. Our results highlight a mechanistic link between NKCC1 expression and behavioral abnormalities in DS mice and establish a molecular target for new therapeutic approaches, including gene therapy, to treat brain disorders characterized by neuronal chloride imbalance.

Mitochondria: new players in homeostatic regulation of firing rate set points

Neural circuit functions are stabilized by homeostatic processes at long timescales in response to changes in behavioral states, experience, and learning. However, it remains unclear which specific physiological variables are being stabilized and which cellular or neural network components compose the homeostatic machinery. At this point, most evidence suggests that the distribution of firing rates among neurons in a neuronal circuit is the key variable that is maintained around a set-point value in a process called 'firing rate homeostasis.' Here, we review recent findings that implicate mitochondria as central players in mediating firing rate homeostasis. While mitochondria are known to regulate neuronal variables such as synaptic vesicle release or intracellular calcium concentration, the mitochondrial signaling pathways that are essential for firing rate homeostasis remain largely unknown. We used basic concepts of control theory to build a framework for classifying possible components of the homeostatic machinery that stabilizes firing rate, and we particularly emphasize the potential role of sleep and wakefulness in this homeostatic process. This framework may facilitate the identification of new homeostatic pathways whose malfunctions drive instability of neural circuits in distinct brain disorders.

Highly active neurons emerging in vitro

Mean firing rates vary across neurons in a neuronal network. Although most neurons infrequently emit spikes, a small fraction of neurons exhibit extremely high frequencies of spikes; this fraction of neurons plays a pivotal role in information processing, however, little is known about how these outliers emerge and whether they are maintained over time. In primary cultures of mouse hippocampal neurons, we traced highly active neurons every 24 h for 7 wk by optically observing the fluorescent protein dVenus; the expression of dVenus was controlled by the promoter of Arc, an immediate early gene that is induced by neuronal activity. Under default-mode conditions, 0.3%-0.4% of neurons were spontaneously Arc-dVenus positive, exhibiting high firing rates. These neurons were spatially clustered, exhibited intermittently repeated dVenus expression, and often continued to express Arc-dVenus for approximately 2 wk. Thus, highly active neurons constitute a few select functional subpopulations in the neuronal network.NEW & NOTEWORTHY The overdispersion of neuronal activity levels can often be attributed to very few neurons exhibiting extremely high firing rates, but due to technical difficulty, no studies have examined how these outliers are selected during development and whether they are maintained over time. We optically monitored highly active neurons for as long as 7 wk in vitro and found that they constituted a unique population that was different from other "mediocre" neurons with normal firing rates.

The brain in motion: How ensemble fluidity drives memory-updating and flexibility

While memories are often thought of as flashbacks to a previous experience, they do not simply conserve veridical representations of the past but must continually integrate new information to ensure survival in dynamic environments. Therefore, 'drift' in neural firing patterns, typically construed as disruptive 'instability' or an undesirable consequence of noise, may actually be useful for updating memories. In our view, continual modifications in memory representations reconcile classical theories of stable memory traces with neural drift. Here we review how memory representations are updated through dynamic recruitment of neuronal ensembles on the basis of excitability and functional connectivity at the time of learning. Overall, we emphasize the importance of considering memories not as static entities, but instead as flexible network states that reactivate and evolve across time and experience.

Functional strengthening through synaptic scaling upon connectivity disruption in neuronal cultures

An elusive phenomenon in network neuroscience is the extent of neuronal activity remodeling upon damage. Here, we investigate the action of gradual synaptic blockade on the effective connectivity in cortical networks in vitro. We use two neuronal cultures configurations-one formed by about 130 neuronal aggregates and another one formed by about 600 individual neurons-and monitor their spontaneous activity upon progressive weakening of excitatory connectivity. We report that the effective connectivity in all cultures exhibits a first phase of transient strengthening followed by a second phase of steady deterioration. We quantify these phases by measuring GEFF, the global efficiency in processing network information. We term hyperefficiency the sudden strengthening of GEFF upon network deterioration, which increases by 20-50% depending on culture type. Relying on numerical simulations we reveal the role of synaptic scaling, an activity-dependent mechanism for synaptic plasticity, in counteracting the perturbative action, neatly reproducing the observed hyperefficiency. Our results demonstrate the importance of synaptic scaling as resilience mechanism.

Impact of Physical Obstacles on the Structural and Effective Connectivity of in silico Neuronal Circuits

Scaffolds and patterned substrates are among the most successful strategies to dictate the connectivity between neurons in culture. Here, we used numerical simulations to investigate the capacity of physical obstacles placed on a flat substrate to shape structural connectivity, and in turn collective dynamics and effective connectivity, in biologically-realistic neuronal networks. We considered μ-sized obstacles placed in mm-sized networks. Three main obstacle shapes were explored, namely crosses, circles and triangles of isosceles profile. They occupied either a small area fraction of the substrate or populated it entirely in a periodic manner. From the point of view of structure, all obstacles promoted short length-scale connections, shifted the in- and out-degree distributions toward lower values, and increased the modularity of the networks. The capacity of obstacles to shape distinct structural traits depended on their density and the ratio between axonal length and substrate diameter. For high densities, different features were triggered depending on obstacle shape, with crosses trapping axons in their vicinity and triangles funneling axons along the reverse direction of their tip. From the point of view of dynamics, obstacles reduced the capacity of networks to spontaneously activate, with triangles in turn strongly dictating the direction of activity propagation. Effective connectivity networks, inferred using transfer entropy, exhibited distinct modular traits, indicating that the presence of obstacles facilitated the formation of local effective microcircuits. Our study illustrates the potential of physical constraints to shape structural blueprints and remodel collective activity, and may guide investigations aimed at mimicking organizational traits of biological neuronal circuits.

The Amyloid Precursor Protein C-Terminal Domain Alters CA1 Neuron Firing, Modifying Hippocampus Oscillations and Impairing Spatial Memory Encoding

There is a growing consensus that Alzheimer's disease (AD) involves failure of the homeostatic machinery, which underlies the firing stability of neural circuits. What are the culprits leading to neuron firing instability? The amyloid precursor protein (APP) is central to AD pathogenesis, and we recently showed that its intracellular domain (AICD) could modify synaptic signal integration. We now hypothesize that AICD modifies neuron firing activity, thus contributing to the disruption of memory processes. Using cellular, electrophysiological, and behavioral techniques, we show that pathological AICD levels weaken CA1 neuron firing activity through a gene-transcription-dependent mechanism. Furthermore, increased AICD production in hippocampal neurons modifies oscillatory activity, specifically in the γ-frequency range, and disrupts spatial memory task. Collectively, our data suggest that AICD pathological levels, observed in AD mouse models and in human patients, might contribute to progressive neuron homeostatic failure, driving the shift from normal aging to AD.

Presynaptic inhibition rapidly stabilises recurrent excitation in the face of plasticity

Hebbian plasticity, a mechanism believed to be the substrate of learning and memory, detects and further enhances correlated neural activity. Because this constitutes an unstable positive feedback loop, it requires additional homeostatic control. Computational work suggests that in recurrent networks, the homeostatic mechanisms observed in experiments are too slow to compensate instabilities arising from Hebbian plasticity and need to be complemented by rapid compensatory processes. We suggest presynaptic inhibition as a candidate that rapidly provides stability by compensating recurrent excitation induced by Hebbian changes. Presynaptic inhibition is mediated by presynaptic GABA receptors that effectively and reversibly attenuate transmitter release. Activation of these receptors can be triggered by excess network activity, hence providing a stabilising negative feedback loop that weakens recurrent interactions on sub-second timescales. We study the stabilising effect of presynaptic inhibition in recurrent networks, in which presynaptic inhibition is implemented as a multiplicative reduction of recurrent synaptic weights in response to increasing inhibitory activity. We show that networks with presynaptic inhibition display a gradual increase of firing rates with growing excitatory weights, in contrast to traditional excitatory-inhibitory networks. This alleviates the positive feedback loop between Hebbian plasticity and network activity and thereby allows homeostasis to act on timescales similar to those observed in experiments. Our results generalise to spiking networks with a biophysically more detailed implementation of the presynaptic inhibition mechanism. In conclusion, presynaptic inhibition provides a powerful compensatory mechanism that rapidly reduces effective recurrent interactions and thereby stabilises Hebbian learning.

Synapses between neurons change during learning and memory formation, a process termed synaptic plasticity. Established models of plasticity rely on strengthening synapses of co-active neurons. In recurrent networks, mutually connected neurons tend to be co-active. The emerging positive feedback loop is believed to be counteracted by homeostatic mechanisms that aim to keep neural activity at a given set point. However, theoretical work indicates that experimentally observed forms of homeostasis are too slow to maintain stable network activity. In this article, we suggest that presynaptic inhibition can alleviate this problem. Presynaptic inhibition is an inhibitory mechanism that weakens synapses rather than suppressing neural activity. Using mathematical analyses and computer simulations, we show that presynaptic inhibition can compensate the strengthening of recurrent connections and thus stabilises neural networks subject to synaptic plasticity, even if homeostasis acts on biologically plausible timescales.

Firing rate homeostasis counteracts changes in stability of recurrent neural networks caused by synapse loss in Alzheimer's disease

The impairment of cognitive function in Alzheimer's disease is clearly correlated to synapse loss. However, the mechanisms underlying this correlation are only poorly understood. Here, we investigate how the loss of excitatory synapses in sparsely connected random networks of spiking excitatory and inhibitory neurons alters their dynamical characteristics. Beyond the effects on the activity statistics, we find that the loss of excitatory synapses on excitatory neurons reduces the network's sensitivity to small perturbations. This decrease in sensitivity can be considered as an indication of a reduction of computational capacity. A full recovery of the network's dynamical characteristics and sensitivity can be achieved by firing rate homeostasis, here implemented by an up-scaling of the remaining excitatory-excitatory synapses. Mean-field analysis reveals that the stability of the linearised network dynamics is, in good approximation, uniquely determined by the firing rate, and thereby explains why firing rate homeostasis preserves not only the firing rate but also the network's sensitivity to small perturbations.

Global sleep homeostasis reflects temporally and spatially integrated local cortical neuronal activity

Sleep homeostasis manifests as a relative constancy of its daily amount and intensity. Theoretical descriptions define ‘Process S’, a variable with dynamics dependent on global sleep-wake history, and reflected in electroencephalogram (EEG) slow wave activity (SWA, 0.5–4 Hz) during sleep. The notion of sleep as a local, activity-dependent process suggests that activity history must be integrated to determine the dynamics of global Process S. Here, we developed novel mathematical models of Process S based on cortical activity recorded in freely behaving mice, describing local Process S as a function of the deviation of neuronal firing rates from a locally defined set-point, independent of global sleep-wake state. Averaging locally derived Processes S and their rate parameters yielded values resembling those obtained from EEG SWA and global vigilance states. We conclude that local Process S dynamics reflects neuronal activity integrated over time, and global Process S reflects local processes integrated over space.

M-Current Inhibition in Hippocampal Excitatory Neurons Triggers Intrinsic and Synaptic Homeostatic Responses at Different Temporal Scales

Persistent alterations in neuronal activity elicit homeostatic plastic changes in synaptic transmission and/or intrinsic excitability. However, it is unknown whether these homeostatic processes operate in concert or at different temporal scales to maintain network activity around a set-point value. Here we show that chronic neuronal hyperactivity, induced by M-channel inhibition, triggered intrinsic and synaptic homeostatic plasticity at different timescales in cultured hippocampal pyramidal neurons from mice of either sex. Homeostatic changes of intrinsic excitability occurred at a fast timescale (1-4 h) and depended on ongoing spiking activity. This fast intrinsic adaptation included plastic changes in the threshold current and a distal relocation of FGF14, a protein physically bridging Na_v1.6 and K_v7.2 channels along the axon initial segment. In contrast, synaptic adaptations occurred at a slower timescale (∼2 d) and involved decreases in miniature EPSC amplitude. To examine how these temporally distinct homeostatic responses influenced hippocampal network activity, we quantified the rate of spontaneous spiking measured by multielectrode arrays at extended timescales. M-Channel blockade triggered slow homeostatic renormalization of the mean firing rate (MFR), concomitantly accompanied by a slow synaptic adaptation. Thus, the fast intrinsic adaptation of excitatory neurons is not sufficient to account for the homeostatic normalization of the MFR. In striking contrast, homeostatic adaptations of intrinsic excitability and spontaneous MFR failed in hippocampal GABAergic inhibitory neurons, which remained hyperexcitable following chronic M-channel blockage. Our results indicate that a single perturbation such as M-channel inhibition triggers multiple homeostatic mechanisms that operate at different timescales to maintain network mean firing rate.SIGNIFICANCE STATEMENT Persistent alterations in synaptic input elicit homeostatic plastic changes in neuronal activity. Here we show that chronic neuronal hyperexcitability, induced by M-type potassium channel inhibition, triggered intrinsic and synaptic homeostatic plasticity at different timescales in hippocampal excitatory neurons. The data indicate that the fast adaptation of intrinsic excitability depends on ongoing spiking activity but is not sufficient to provide homeostasis of the mean firing rate. Our results show that a single perturbation such as M-channel inhibition can trigger multiple homeostatic processes that operate at different timescales to maintain network mean firing rate.

Assessment of Spontaneous Neuronal Activity In Vitro Using Multi-Well Multi-Electrode Arrays: Implications for Assay Development

Multi-electrode arrays (MEAs) are being more widely used by researchers as an instrument platform for monitoring prolonged, non-destructive recordings of spontaneously firing neurons in vitro for applications in modeling Alzheimer's, Parkinson's, schizophrenia, and many other diseases of the human CNS. With the more widespread use of these instruments, there is a need to examine the prior art of studies utilizing MEAs and delineate best practices for data acquisition and analysis to avoid errors in interpretation of the resultant data. Using a dataset of recordings from primary rat (Rattus norvegicus) cortical cultures, methods and statistical power for discerning changes in neuronal activity on the array level are examined. Further, a method for unsupervised spike sorting is implemented, allowing for the resolution of action potential incidents down to the single neuron level. Following implementation of spike sorting, the dynamics of firing frequency across populations of individual neurons and networks are examined longitudinally. Finally, the ability to detect a frequency independent phenotype, the change in action potential amplitude, is demonstrated through the use of pore-forming neurotoxin treatments. Taken together, this study provides guidance and tools for users wishing to incorporate multi-well MEA usage into their studies.

Differential Effects of Open- and Closed-Loop Intracortical Microstimulation on Firing Patterns of Neurons in Distant Cortical Areas

Intracortical microstimulation can be used successfully to modulate neuronal activity. Activity-dependent stimulation (ADS), in which action potentials recorded extracellularly from a single neuron are used to trigger stimulation at another cortical location (closed-loop), is an effective treatment for behavioral recovery after brain lesion, but the related neurophysiological changes are still not clear. Here, we investigated the ability of ADS and random stimulation (RS) to alter firing patterns of distant cortical locations. We recorded 591 neuronal units from 23 Long-Evan healthy anesthetized rats. Stimulation was delivered to either forelimb or barrel field somatosensory cortex, using either RS or ADS triggered from spikes recorded in the rostral forelimb area (RFA). Both RS and ADS stimulation protocols rapidly altered spike firing within RFA compared with no stimulation. We observed increase in firing rates and change of spike patterns. ADS was more effective than RS in increasing evoked spikes during the stimulation periods, by producing a reliable, progressive increase in stimulus-related activity over time and an increased coupling of the trigger channel with the network. These results are critical for understanding the efficacy of closed-loop electrical microstimulation protocols in altering activity patterns in interconnected brain networks, thus modulating cortical state and functional connectivity.

Firing Rate Homeostasis Can Occur in the Absence of Neuronal Activity-Regulated Transcription

Despite dynamic inputs, neuronal circuits maintain relatively stable firing rates over long periods. This maintenance of firing rate, or firing rate homeostasis, is likely mediated by homeostatic mechanisms such as synaptic scaling and regulation of intrinsic excitability. Because some of these homeostatic mechanisms depend on transcription of activity-regulated genes, including Arc and Homer1a, we hypothesized that activity-regulated transcription would be required for firing rate homeostasis. Surprisingly, however, we found that cultured mouse cortical neurons from both sexes grown on multi-electrode arrays homeostatically adapt their firing rates to persistent pharmacological stimulation even when activity-regulated transcription is disrupted. Specifically, we observed firing rate homeostasis in Arc knock-out neurons, as well as knock-out neurons lacking the activity-regulated transcription factors AP1 and SRF. Firing rate homeostasis also occurred normally during acute pharmacological blockade of transcription. Thus, firing rate homeostasis in response to increased neuronal activity can occur in the absence of neuronal-activity-regulated transcription.SIGNIFICANCE STATEMENT Neuronal circuits maintain relatively stable firing rates even in the face of dynamic circuit inputs. Understanding the molecular mechanisms that enable this firing rate homeostasis could potentially provide insight into neuronal diseases that present with an imbalance of excitation and inhibition. It has long been proposed that activity-regulated transcription could underlie firing rate homeostasis because activity-regulated genes turn on when neurons are above their target firing rates and include many genes that could regulate firing rate. Surprisingly, despite this prediction, we found that cortical neurons can undergo firing rate homeostasis in the absence of activity-regulated transcription, indicating that firing rate homeostasis can be controlled by non-transcriptional mechanisms.

Machine learning-guided channelrhodopsin engineering enables minimally invasive optogenetics

We engineered light-gated channelrhodopsins (ChRs) whose current strength and light sensitivity enable minimally-invasive neuronal circuit interrogation. Current ChR tools applied to the mammalian brain require intracranial surgery for transgene delivery and implantation of invasive fiber-optic cables to produce light-dependent activation of a small volume of tissue. To facilitate expansive optogenetics without the need for invasive implants, our engineering approach leverages the significant literature of ChR variants to train statistical models for the design of new, high-performance ChRs. With Gaussian Process models trained on a limited experimental set of 102 functionally characterized ChRs, we designed high-photocurrent ChRs with unprecedented light sensitivity; three of these, ChRger1–3, enable optogenetic activation of the nervous system via minimally-invasive systemic transgene delivery, not possible previously due to low per-cell transgene copy produced by systemic delivery. ChRger2 enables light-induced neuronal excitation without invasive intracranial surgery for virus delivery or fiber optic implantation, i.e. enables minimally-invasive optogenetics.

Cortical Circuit Dynamics Are Homeostatically Tuned to Criticality In Vivo

Homeostatic mechanisms stabilize neuronal activity in vivo, but whether this process gives rise to balanced network dynamics is unknown. Here, we continuously monitored the statistics of network spiking in visual cortical circuits in freely behaving rats for 9 days. Under control conditions in light and dark, networks were robustly organized around criticality, a regime that maximizes information capacity and transmission. When input was perturbed by visual deprivation, network criticality was severely disrupted and subsequently restored to criticality over 48 h. Unexpectedly, the recovery of excitatory dynamics preceded homeostatic plasticity of firing rates by >30 h. We utilized model investigations to manipulate firing rate homeostasis in a cell-type-specific manner at the onset of visual deprivation. Our results suggest that criticality in excitatory networks is established by inhibitory plasticity and architecture. These data establish that criticality is consistent with a homeostatic set point for visual cortical dynamics and suggest a key role for homeostatic regulation of inhibition.

Unique Features of Network Bursts Emerge From the Complex Interplay of Excitatory and Inhibitory Receptors in Rat Neocortical Networks

Spontaneous network activity plays a fundamental role in the formation of functional networks during early development. The landmark of this activity is the recurrent emergence of intensive time-limited network bursts (NBs) rapidly spreading across the entire dissociated culture in vitro. The main excitatory mediators of NBs are glutamatergic alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptors (AMPARs) and N-Methyl-D-aspartic-acid receptors (NMDARs) that express fast and slow ion channel kinetics, respectively. The fast inhibition of the activity is mediated through gamma-aminobutyric acid type A receptors (GABA_ARs). Although the AMPAR, NMDAR and GABA_AR kinetics have been biophysically characterized in detail at the monosynaptic level in a variety of brain areas, the unique features of NBs emerging from the kinetics and the complex interplay of these receptors are not well understood. The goal of this study is to analyze the contribution of fast GABA_ARs on AMPAR- and NMDAR- mediated spontaneous NB activity in dissociated neonatal rat cortical cultures at 3 weeks in vitro. The networks were probed by both acute and gradual application of each excitatory receptor antagonist and combinations of acute excitatory and inhibitory receptor antagonists. At the same time, the extracellular network-wide activity was recorded with microelectrode arrays (MEAs). We analyzed the characteristic NB measures extracted from NB rate profiles and the distributions of interspike intervals, interburst intervals, and electrode recruitment time as well as the similarity of spatio-temporal patterns of network activity under different receptor antagonists. We show that NBs were rapidly initiated and recruited as well as diversely propagated by AMPARs and temporally and spatially maintained by NMDARs. GABA_ARs reduced the spiking frequency in AMPAR-mediated networks and dampened the termination of NBs in NMDAR-mediated networks as well as slowed down the recruitment of activity in all networks. Finally, we show characteristic super bursts composed of slow NBs with highly repetitive spatio-temporal patterns in gradually AMPAR blocked networks. To the best of our knowledge, this study is the first to unravel in detail how the three main mediators of synaptic transmission uniquely shape the NB characteristics, such as the initiation, maintenance, recruitment and termination of NBs in cortical cell cultures in vitro.

Mom Controls the Thermostat: Mitochondria Influence the Neuronal Firing Set Point

Mitochondrial Regulation of the Hippocampal Firing Rate Set Point and Seizure Susceptibility

Styr B, Gonen N, Zarhin D, Ruggiero A, Atsmon R, Gazit N, Braun G, Frere S, Vertkin I, Shapira I, Harel M, Heim LR, Katsenelson M, Rechnitz O, Fadila S, Derdikman D, Rubinstein M, Geiger T, Ruppin E, Slutsky I. Neuron. 2019. pii: S0896-6273(19)30334-4. doi:10.1016/j.neuron.2019.03.045. [Epub ahead of print] PMID: 31047779.

Maintaining average activity within a set-point range constitutes a fundamental property of central neural circuits. However, whether and how activity set points are regulated remains unknown. Integrating genome-scale metabolic modeling and experimental study of neuronal homeostasis, we identified mitochondrial dihydroorotate dehydrogenase (DHODH) as a regulator of activity set points in hippocampal networks. The DHODH inhibitor teriflunomide stably suppressed mean firing rates via synaptic and intrinsic excitability mechanisms by modulating mitochondrial Ca2+ buffering and spare respiratory capacity. Bidirectional activity perturbations under DHODH blockade triggered firing rate compensation, while stabilizing firing to the lower level, indicating a change in the firing rate set point. In vivo, teriflunomide decreased CA3–CA1 synaptic transmission and CA1 mean firing rate and attenuated susceptibility to seizures, even in the intractable Dravet syndrome epilepsy model. Our results uncover mitochondria as a key regulator of activity set points, demonstrate the differential regulation of set points and compensatory mechanisms, and propose a new strategy to treat epilepsy.

Sensitivity to Stimulus Irregularity Is Inherent in Neural Networks

Behavior is controlled by complex neural networks in which neurons process thousands of inputs. However, even short spike trains evoked in a single cortical neuron were demonstrated to be sufficient to influence behavior in vivo. Specifically, irregular sequences of interspike intervals (ISIs) had a more reliable influence on behavior despite their resemblance to stochastic activity. Similarly, irregular tactile stimulation led to higher rates of behavioral responses. In this study, we identify the mechanisms enabling this sensitivity to stimulus irregularity (SSI) on the neuronal and network levels using simulated spiking neural networks. Matching in vivo experiments, we find that irregular stimulation elicits more detectable network events (bursts) than regular stimulation. Dissecting the stimuli, we identify short ISIs-occurring more frequently in irregular stimulations-as the main drivers of SSI rather than complex irregularity per se. In addition, we find that short-term plasticity modulates SSI. We subsequently eliminate the different mechanisms in turn to assess their role in generating SSI. Removing inhibitory interneurons, we find that SSI is retained, suggesting that SSI is not dependent on inhibition. Removing recurrency, we find that SSI is retained due to the ability of individual neurons to integrate activity over short timescales ("cell memory"). Removing single-neuron dynamics, we find that SSI is retained based on the short-term retention of activity within the recurrent network structure ("network memory"). Finally, using a further simplified probabilistic model, we find that local network structure is not required for SSI. Hence, SSI is identified as a general property that we hypothesize to be ubiquitous in neural networks with different structures and biophysical properties. Irregular sequences contain shorter ISIs, which are the main drivers underlying SSI. The experimentally observed SSI should thus generalize to other systems, suggesting a functional role for irregular activity in cortex.

Alzheimer's Disease: From Firing Instability to Homeostasis Network Collapse

Alzheimer's disease (AD) starts from pure cognitive impairments and gradually progresses into degeneration of specific brain circuits. Although numerous factors initiating AD have been extensively studied, the common principles underlying the transition from cognitive deficits to neuronal loss remain unknown. Here we describe an evolutionarily conserved, integrated homeostatic network (IHN) that enables functional stability of central neural circuits and safeguards from neurodegeneration. We identify the critical modules comprising the IHN and propose a central role of neural firing in controlling the complex homeostatic network at different spatial scales. We hypothesize that firing instability and impaired synaptic plasticity at early AD stages trigger a vicious cycle, leading to dysregulation of the whole IHN. According to this hypothesis, the IHN collapse represents the major driving force of the transition from early memory impairments to neurodegeneration. Understanding the core elements of homeostatic control machinery, the reciprocal connections between distinct IHN modules, and the role of firing homeostasis in this hierarchy has important implications for physiology and should offer novel conceptual approaches for AD and other neurodegenerative disorders.

Mitochondrial Regulation of the Hippocampal Firing Rate Set Point and Seizure Susceptibility

Maintaining average activity within a set-point range constitutes a fundamental property of central neural circuits. However, whether and how activity set points are regulated remains unknown. Integrating genome-scale metabolic modeling and experimental study of neuronal homeostasis, we identified mitochondrial dihydroorotate dehydrogenase (DHODH) as a regulator of activity set points in hippocampal networks. The DHODH inhibitor teriflunomide stably suppressed mean firing rates via synaptic and intrinsic excitability mechanisms by modulating mitochondrial Ca²⁺ buffering and spare respiratory capacity. Bi-directional activity perturbations under DHODH blockade triggered firing rate compensation, while stabilizing firing to the lower level, indicating a change in the firing rate set point. In vivo, teriflunomide decreased CA3-CA1 synaptic transmission and CA1 mean firing rate and attenuated susceptibility to seizures, even in the intractable Dravet syndrome epilepsy model. Our results uncover mitochondria as a key regulator of activity set points, demonstrate the differential regulation of set points and compensatory mechanisms, and propose a new strategy to treat epilepsy.

Confocal Imaging and k-Means Clustering of GABAB and mGluR Mediated Modulation of Ca2+ Spiking in Hippocampal Neurons

Imaging cytosolic calcium in neurons is emerging as a new tool in neurological disease diagnosis, drug screening, and toxicity testing. Ca²⁺ oscillation signatures show a significant variation depending on GPCR targeting agonists. Quantification of Ca²⁺ spike trains in ligand induced Ca²⁺ oscillations remains challenging due to their inherent heterogeneity in primary culture. Moreover, there is no framework available for identification of optimal number of clusters and distance metric to cluster Ca²⁺ spike trains. Using quantitative confocal imaging and clustering analysis, we show the characterization of Ca²⁺ spiking in GPCR targeting drug-treated primary culture of hippocampal neurons. A systematic framework for selection of the clustering method instead of an intuition-based method was used to optimize the cluster number and distance metric. The results discern neurons with diverse Ca²⁺ response patterns, including higher amplitude fast spiking and lower spiking responses, and their relative percentage in a neuron population in absence and presence of GPCR-targeted drugs. The proposed framework was employed to show that the  clustering pattern of Ca²⁺ spiking can be controlled using GABA_B and mGluR targeting drugs. This approach can be used for unbiased measurement of neural activity and identification of spiking population with varying amplitude and frequencies, providing a platform for high-content drug screening.

Neuroengineering Tools For Studying The Effect Of Intracortical Microstimulation In Rodent Models

Intracortical microstimulation can be successfully used to manipulate neuronal activity and connectivity, thus representing a potentially powerful tool to steer neuroplasticity occurring after brain injury. Activity-dependent stimulation (ADS), in which the spikes recorded from a single neuron are used to trigger stimulation at another cortical location, is able to potentiate cortical connections between distant brain areas. Here, we developed an experimental procedure and a computational pipeline aimed at investigating the ability of ADS to induce changes in intra-cortical activity of healthy anesthetized rats.

Single-layer graphene modulates neuronal communication and augments membrane ion currents

The use of graphene-based materials to engineer sophisticated biosensing interfaces that can adapt to the central nervous system requires a detailed understanding of how such materials behave in a biological context. Graphene's peculiar properties can cause various cellular changes, but the underlying mechanisms remain unclear. Here, we show that single-layer graphene increases neuronal firing by altering membrane-associated functions in cultured cells. Graphene tunes the distribution of extracellular ions at the interface with neurons, a key regulator of neuronal excitability. The resulting biophysical changes in the membrane include stronger potassium ion currents, with a shift in the fraction of neuronal firing phenotypes from adapting to tonically firing. By using experimental and theoretical approaches, we hypothesize that the graphene-ion interactions that are maximized when single-layer graphene is deposited on electrically insulating substrates are crucial to these effects.

Homeostatic plasticity in neural development

Throughout life, neural circuits change their connectivity, especially during development, when neurons frequently extend and retract dendrites and axons, and form and eliminate synapses. In spite of their changing connectivity, neural circuits maintain relatively constant activity levels. Neural circuits achieve functional stability by homeostatic plasticity, which equipoises intrinsic excitability and synaptic strength, balances network excitation and inhibition, and coordinates changes in circuit connectivity. Here, we review how diverse mechanisms of homeostatic plasticity stabilize activity in developing neural circuits.

In vitro Cortical Network Firing is Homeostatically Regulated: A Model for Sleep Regulation

Prolonged wakefulness leads to a homeostatic response manifested in increased amplitude and number of electroencephalogram (EEG) slow waves during recovery sleep. Cortical networks show a slow oscillation when the excitatory inputs are reduced (during slow wave sleep, anesthesia), or absent (in vitro preparations). It was recently shown that a homeostatic response to electrical stimulation can be induced in cortical cultures. Here we used cortical cultures grown on microelectrode arrays and stimulated them with a cocktail of waking neuromodulators. We found that recovery from stimulation resulted in a dose-dependent homeostatic response. Specifically, the inter-burst intervals decreased, the burst duration increased, the network showed higher cross-correlation and strong phasic synchronized burst activity. Spectral power below <1.75 Hz significantly increased and the increase was related to steeper slopes of bursts. Computer simulation suggested that a small number of clustered neurons could potently drive the behavior of the network both at baseline and during recovery. Thus, this in vitro model appears valuable for dissecting network mechanisms of sleep homeostasis.