Frontiers in Neural Circuits

Recent studies have shown that symbiotic bacteria can have drastic effects on host neurobiology, but few simple, accessible models currently exist in which to study these interactions. Hawaiian bobtail squid (Euprymna scolopes) participate in a binary symbiosis with the bacterium Vibrio fischeri, a population of which resides in a specialized hindgut-derived organ called the light organ. Upon colonization by V. fischeri, the light organ undergoes transcriptional changes that suggest neurons are impacted by the initiation of symbiosis, but the nascent light organs innervation has remained uncharacterized. Here, we show that the light organ-associated nervous system (LONS) in hatchling E. scolopes is a remarkably complex segment of the peripheral nervous system. The LONS is largely plexiform and originates from two primary nerves connected by a local commissure. The abundance of synapsin-like immunoreactivity (-lir) indicates that the lobe plexus is highly interconnected. We also highlight a small number of serotonin-lir neurites that innervate the anterior appendages whose developmental fate may be directly affected by symbiont-driven light organ morphogenesis. Finally, we present evidence that a limited but diverse population of neurons reside within the light organ and are often located near internal symbiont-interacting structures. This description of the E. scolopes LONS serves to provide a foundation from which to investigate how beneficial bacterial symbionts affect host peripheral neurobiology in a tractable model system.

Understanding how minimal perturbations influence cortical network dynamics remains a central challenge in neural engineering. While single-cell stimulation has been shown to affect population activity, the resulting variability is often treated as noise rather than an informative feature of network behavior. Here, we investigated how single-cell stimulation reflects heterogeneous modulation of neuronal recruitment and the extent to which these effects can be explained by the functional state of the stimulated cell. For this, we combined single-cell optogenetic stimulation with wide-field calcium imaging in cortical cultures. In each network, a single stimulation event was induced, and subsequent alterations in stimulus-coupled recruitment, synchrony, and pairwise correlations were quantified. Additionally, we evaluated whether the baseline functional state of the stimulated neurons, including their event activity levels and Pearson correlation structures, were linked to the observed network responses. Single cell stimulation induced effects were transient, and the network dynamics recovered over a few seconds within the responder population. Importantly, Our findings demonstrate that the observed direction and magnitude of recruitment changes were not significantly explained by the functional state of the stimulated neurons, indicating that these parameters do not capture the determinants of perturbation-induced network responses. This highlights a possibilities in the future approaches for characterizing network responsiveness and suggests that additional unobserved features govern the response of microcircuits to localized inputs.

The phenomenon of splitting was originally observed in hamsters which, after prolonged exposure to constant light, exhibit two rest/wake cycles within a subjective day. Splitting is a consequence of the left and right suprachiasmatic nuclei (SCN) falling out of synchrony. While it is known that split activity is characterized by an antiphase relationship between the left and right SCN and between the core and shell within each hemisphere, the role of the commissural projections that connect the right and left SCN is not known. In the present study, we investigate the impact of the inter-hemispheric connections on the split and unsplit dynamics of a computational model of the bilateral SCN. Our model has 4 nodes corresponding to each right and left core and shell. We simulated our bilateral model under different lighting conditions and measured its period and the phase relationships among the 4 nodes. To further characterize the dynamics of the system, we performed a bifurcation analysis. We found that the bilateral model automatically splits unless entrained by bright light/dark cycles, or unless it has excitatory inter-hemispheric connections. This suggests that excitatory cross-connections may be important for freerunning behavior. We found that constant light of varying intensities transitions the model between split and unsplit activity only in very limited conditions, but the strength and polarity of the contralateral connections play a much greater role in this dynamical transition. These findings suggest that splitting may involve plasticity of the inter-hemispheric connections of the SCN.

Electrical transmission is mediated by intercellular channels that cluster into structures known as gap junctions (GJ). In vertebrates, GJ channels are encoded by the gene family of connexin (Cx) proteins that assemble as hexamers, termed hemichannels, in the pre- and postsynaptic membranes, and that subsequently dock to form GJ channels. Auditory contacts on the fish Mauthner cells serve as model to study the properties and organization of vertebrate electrical synapses. Electrical transmission at these synapses is mediated by multiple co-existing GJs at which the presence of intercellular channels is regulated by a molecular scaffold. Zebrafish contain four homologs of the neuronal Cx36: Cx35.5 and Cx35.1 (gjd2a and b, respectively), and Cx34.1 and Cx34.7 (gjd1a and b). Cx mutations suggested that GJs are formed by heterotypic channels made of presynaptic Cx35.5 and postsynaptic Cx34.1. Using transgenic fish in which Cxs were tagged, we found that a second Cx, Cx34.7, is present together with Cx34.1 on the postsynaptic side at some but not all GJs at these terminals. When exogenously expressed, both Cx34.1 and Cx34.7 formed heterotypic functional channels with Cx35.5, each with substantially different voltage-dependent properties, indicating they can serve differential functions. However, we previously demonstrated that electrical transmission is lost in Cx34.1 but not Cx34.7 null mutants, suggesting that Cx34.7 cannot compensate for the loss of Cx34, despite the intrinsic ability of Cx34.1 and Cx34.7 to create functional channels. The findings reveal an unanticipated functional organization in the electrical synapse, where Cx34.1 is obligatory and Cx34.7 accessory, roles that appear to be defined by the postsynaptic molecular scaffold, with two postsynaptic Cxs possibly assembling under specific functional contexts. Thus, our results indicate that electrical synapses share an organizational motif with chemical synapses, akin to how they combine postsynaptic receptor types to modify synaptic function.

Molecular layer interneurons (MLIs) play a crucial role in modulating the output of the cerebellar cortex through their inhibition of Purkinje cells. MLIs also inhibit other MLIs synaptically and are coupled electrically through gap junctions. While synchronization of MLIs has been observed, comprehensive understanding of the role of gap junctional coupling in shaping MLI network activity is lacking. Dendro-dendritic gap junctional coupling in MLIs involves propagation of signals to and from the dendritic gap junction location which can lead to neural synchronization. However, how this is regulated by the intrinsic electrical properties of MLIs, including dendritic properties, is poorly understood. In this study, we apply conductance-based computational modelling to examine the effect of dendritic filtering on gap junctional coupling in pairs of ball-and-stick MLI models, demonstrating that gap junctional properties, rather than the active dendritic properties of MLIs, primarily dictate gap junction-driven synchronization. By systematically reducing the ball-and-stick model to a one-compartment MLI model, we additionally investigate the role of MLI gap junctional coupling in mediating MLI network synchrony. Our results reveal that transient AMPA input drives brief network-wide synchronization, whereas NMDA-mediated elevated firing enables gap junction-dependent oscillatory synchronization that is further enhanced by MLI-MLI inhibition in a positive feedback loop, producing pronounced peaks of network coactivity resembling sensory-evoked MLI activity observed in vivo. These findings provide important insights into network dynamics of MLIs and how gap junctions shape their activity, with broader implications for other neural networks that rely on gap junctional coupling.

Repetitive transcranial magnetic stimulation (rTMS) is used widely in neuroscience to study and alter neural plasticity. The cellular mechanisms underlying the effect of rTMS on the brain remain unclear but is primarily thought to act via activity-dependent synaptic plasticity mechanisms. Here we investigated whether chronic repetitive magnetic stimulation in vitro and in vivo can induce another form of activity-dependent neural plasticity, axon initial segment (AIS) plasticity. Cortical neurons isolated from postnatal wild-type mice were stimulated with 6 hours of sham, repetitive magnetic stimulation in the form of intermittent theta-burst stimulation (iTBS), or 15 mM potassium chloride, with changes to AIS location and length measured +0 hours and +24 hours post-stimulation. In addition, adult transgenic mice expressing green fluorescent protein at the AIS received daily sham or iTBS over the primary motor cortices for 7 consecutive days and processed for microscopy 3 hours after the last stimulation. Analysis of neurons stimulated in vitro showed that chronic iTBS caused bidirectional and time-dependent shifts to the AIS position relative to the soma and a delayed shortening of the AIS length at +24 hours. In the adult mice, 7 consecutive days of daily iTBS decreased AIS lengths in layers 2/3 and 5 pyramidal neurons. Our findings provide in vitro and in vivo evidence that rTMS induces neuronal plasticity outside of the synapse, which may contribute to the long-lasting effect of rTMS on the brain with repeated stimulation protocols.

Hippocampal area CA2 has recently emerged as a critical region for social recognition memory. Furthermore, this understudied region has been implicated in psychiatric diseases and neurodegenerative diseases. There has been accumulating evidence indicating that the pyramidal neurons (PNs) in area CA2 exhibit functional specializations that correlate with somatic position in stratum pyramidale (sp). In this study, we investigated the morphological differences in dendritic architecture of CA2 PNs with a focus on the radial gradient, i.e., along the deep-superficial axis of the sp. We conducted a comprehensive morphological analysis including Sholl intersection profiles, branching order distributions, root angle distributions, and dendritic cable lengths. We found that CA2 PNs have fewer oblique dendrites and a larger number of tuft-like dendrites as compared to CA1 PNs. Furthermore, within the CA2 population, we found that many of the dendritic structural features gradually changed along the radial axis from deep to superficial somatic location, indicating a continuum of dendritic morphology rather than two sharply defined subtypes of pyramidal neurons. This morphological characterization may serve as a starting point to better understand the corresponding functional organization of CA2. The gradual difference between deeper and superficial CA2 PNs suggests a continuum of their computational capabilities beyond two binary functional classes. In briefUsing several methods, we examine the dendritic morphology of over 130 CA2 and CA1 pyramidal neurons and find that many properties such as the cable length and terminal numbers of the dendritic arbors vary as a with the location of the soma in the pyramidal layer. HighlightsO_LIWe use scholl analysis, graph theory and machine learning techniques to quantify the different dendritic morphologies of CA2 pyramidal neurons. C_LIO_LIMany properties of CA2 pyramidal neuron apical dendrites vary as a function of somatic location in the pyramidal layer. C_LIO_LIMore superficial CA2 pyramidal neurons have longer oblique apical dendrites, and shorter tuft dendrites. C_LI

Homeostatic plasticity of intrinsic excitability (IE) in the visual system has been essentially shown at the cortical level but whether thalamic nuclei also express homeostatic plasticity of IE is unknown. We show here that 4 days of monocular deprivation (MD) at eye opening induces a homeostatic change in IE in dorsal lateral geniculate nucleus (dLGN) neurons. Neurons recorded in the dLGN region activated by the deprived eye are more excitable than neurons recorded in the dLGN region activated by the open eye. No significant changes were observed following 7 days of MD, however. Enhanced excitability in neurons from the deprived side after 4 days of MD was associated with a reduced Kv1-dependent LTP-IE, a smaller voltage ramp, and a reduced inter-spike interval, suggesting that Kv1 channels are down-regulated in deprived dLGN neurons. Furthermore, the ankyrin G signal of the axon initial segment was larger in deprived dLGN neurons compared with open ones, indicating that Nav1 channel number also undergoes homeostatic regulation, and Kv1.1 channel signals were lower in deprived neurons compared to open ones. In addition, electrical coupling was found to be strengthened in neurons displaying enhanced IE following either brief (4 days) or long (10 days) MD. These results suggest that homeostatic and Hebbian plasticity in the dLGN share common expression mechanisms involving the regulation of Kv1 channels, Nav1 channels and electrical coupling between relay neurons.

The hippocampus participates in crucial functions such as memory consolidation, spatial processing and emotional regulation that require diverse input from multiple cortical areas that is funneled through the upper layers of the entorhinal cortex (EC), mostly from layer II to the dentate gyrus (DG). Traditional models described 200,000 EC layer II neurons projecting to 1 million granule cells (GCs) in the rat, rendering low divergence (1:5), with each EC neuron establishing about 18,000 synapses with GCs and each GC receiving about 4,000 synapses from EC neurons. In this manuscript, we update this model of connectivity incorporating new features described in the last three decades that include updated populations of EC layer II neurons obtained with design-based stereology, a revised definition of EC layer II based on molecular criteria and selecting reelin expressing neurons as the only layer II neurons projecting to the hippocampus. The updated model shows [~]80,000 neurons from EC layer II projecting to the DG, [~]45,000 from the medial entorhinal cortex (MEC) and [~]35,000 from the lateral entorhinal cortex (LEC) with high divergence of 1:20 and 1:30. We also show that EC layer II neurons may establish [~]90,000-115,000 synapses on GCs, while GCs receive about 8,000 synapses from EC layer II neurons. We estimate a [~]25% redundancy in the connectivity, so each EC neuron may contact [~]68,000-86,000 GCs and each GC would be contacted by [~]3000 neurons from MEC and 3,000 from LEC. In addition, we quantitatively assess a potential projection of mossy cells to the medial molecular layer described in mice, which could have a potential impact on GC inhibition. Overall, we produced a detailed, complete, and updated quantitative model of EC projections to the DG that reveals a much more divergent and richer projection than previously described, with implications for functional models (e.g.: pattern separation) and more widely for building realistic hippocampal models or establishing comparisons across species.

The cerebellar nuclei form the main output structures of the cerebellum and are composed of a deeply conserved set of cell types. Two excitatory cell classes, Class-A and -B, are present in each cerebellar nucleus and mediate all excitatory output of the cerebellum. To provide genetic access to these cell types, here we identified Acan as a marker gene for Class-B cells and generated a knock-in Acan-P2A-Cre mouse line. We demonstrate that this Acan-Cre line selectively labels Class-B neurons in the cerebellar nuclei and validate its use in viral projection tracing. This new mouse line provides a valuable genetic tool to study cerebellar nuclei organization and function.

Dendritic arbor morphology is shaped in part by interactions with neighboring dendrites, and its geometry strongly influences the spatial distribution and strength of synapses. These observations raise the possibility that local dendritic contacts help determine where synapses accumulate and strengthen. Previous work in cultured hippocampal neurons showed that dendrite-dendrite contact sites are non-random and associated with local synaptic clustering. Here we asked whether a different type of dendritic contact, formed between a dendrite and the soma of a neighboring neuron, behaves similarly. Using dissociated hippocampal cultures, immunofluorescence imaging, time-lapse microscopy, quantitative image analysis, stochastic spatial simulations, and minimal quantitative modeling, we identified three recurrent classes of dendrite-soma interactions (DSIs): dendrites crossing directly over a neighboring soma, growing tangentially along the soma perimeter, or contacting the proximal region where a neighboring dendrite emerges from the soma. These interactions were abundant, occurred exclusively between different neurons, and showed substantial structural persistence over several days. Their overall frequency exceeded stochastic predictions across culture densities, and two configurations - proximal and tangential contacts - were selectively enriched above random expectation, whereas soma-crossing contacts were largely consistent with stochastic overlap. DSI composition also changed over development, with proximal contacts becoming progressively more prevalent. At DSI sites, synaptophysin-positive puncta were significantly denser and more intense than on non-interacting dendritic segments, consistent with local enrichment and strengthening of presynaptic specializations. Minimal modeling further indicated that biased formation together with developmental stabilization explains the observed organization better than stochastic geometry alone. These findings identify DSIs as non-random structural motifs in cultured hippocampal networks and suggest that dendrite contact geometry can contribute to synaptic distribution and strengthening.

Short-term plasticity (STP) is the transient fluctuation of connection strength between two neurons depending on the recent history of neuronal activity. STP shapes neurotransmission over time and plays important roles in circuit computations. It is classically quantified ex vivo either at the synaptic level or at the level of spike transmission from the presynaptic neuron to the postsynaptic neuron. However, the exact relationship between the postsynaptic dendritic responses and spike transmission during STP still remains unclear in vivo. Here, we characterized together the STP of both postsynaptic dendritic responses, measured by the postsynaptic field potential (PFP), and spike transmission at the retinocollicular pathway of mice. We found mostly facilitating STP, where the second presynaptic spike occurring within 25 ms induces a larger PFP, and consequently a higher postsynaptic firing rate compared to responses from the first presynaptic spike. Both PFP and spike transmission exhibit short-term facilitation, but to a different degree, where the facilitation in the spike transmission is larger than the PFP. The PFP and spike transmission also exhibit facilitation of different decay time constants, indicating a nonlinear relationship between the two. Interestingly, a preceding postsynaptic spike can induce a similarly large but longer lasting facilitation on the spike transmission upon receiving a subsequent presynaptic input. However, STP of the PFP does not depend on the preceding postsynaptic spike, suggesting that this longer postsynaptic facilitation has a nonsynaptic origin. Overall, our results indicate that STP of the retinocollicular pathway exhibits three different stages: 1) a weak synaptic facilitation of postsynaptic dendritic responses, 2) a strong synaptic facilitation of spike transmission, and 3) a longer lasting nonsynaptic facilitation of spike transmission. Using a computational model, we show that the second STP stage is a direct inheritance from the first STP stage, whereas two opposing nonsynaptic mechanisms with different time constants are needed for the emergence of the third STP stage. These findings provide direct evidence that synaptic and nonsynaptic STPs coexist in vivo, paving the way for large-scale measurement of these STPs and offering a means to monitor the transmission of information in neural circuits of behaving animals.

To avoid image blurring, the vestibulo-ocular (VOR) and the optokinetic (OKR) reflexes stabilize gaze. In all vertebrates, the VOR is mediated via direct projections from the vestibular nuclei to the motor nuclei that control the extraocular muscles. Lampreys show three vestibular nuclei that are well characterized in terms of their projections and sensory inputs, but much less is known about their inputs from other brain regions and the connectivity between them. Using tracer injections and electrophysiological recordings, we show that the lamprey vestibular nuclei are largely interconnected, while their inputs from other brain regions are scarce. The main rostral areas projecting to the vestibular nuclei are the pretectum and the ventral tier of the thalamus, which send ipsilateral inputs to the three vestibular nuclei.

A major outstanding question in neuroscience is whether the neocortex uses the same powerful learning algorithm as current AI models: error backpropagation. One way this could be accomplished is as a function of the temporal derivative (i.e., differences in neural activity states over time), which can closely approximate the backpropagated error gradient. We tested the hypothesis that the direction of synaptic plasticity is a function of the temporal derivative in synaptic activity over the course of a 200 ms (5 Hz) theta cycle. Using mouse hippocampal slices, we drove presynaptic activity across the two 100 ms halves of a 200 ms window at either 25 Hz or 50 Hz, combined with corresponding low and high magnitudes of postsynaptic depolarization, testing all four 2x2 combinations of these low and high activity levels, while measuring the resulting effects on synaptic efficacy (as measured by EPSP amplitude to standard test probes). Consistent with the computational hypothesis, a positive temporal derivative (low to high) resulted in LTP (increased synaptic strength), while a negative temporal derivative (high to low) resulted in LTD. Critically, both no-change conditions (stable low or high across 200 ms) resulted in no net synaptic change, even though the high no-change condition had the highest overall synaptic activity levels. Possible biochemical mechanisms that could support these results are discussed.

Although the cerebellar microcircuit is among the most well-characterized systems for studying neural computation, recent connectomic analyses highlight deviations from canonical models, raising fundamental questions about connectivity, function, and learning in the system. A key feature of the circuit is the convergence of a vast number of parallel fibers (PFs) onto Purkinje cells (PCs). Prevailing models of cerebellar computation assume all-to-all connectivity at this intersection, whereby a single PC "samples" from all PFs. However, experimental evidence suggests that each PC is innervated by only a subset of accessible PFs. This partial sampling creates differential connectivity that may reflect an organizational principle of cerebellar learning, but its implications are poorly understood. Based on electron microscopy (EM) reconstructions, we show that PF innervation of PCs is largely consistent with a Bernoulli model where connections are randomly and independently distributed within anatomical constraints. In further support of a random model, we observe that connections of the ascending branches of granule cells are not predictive of connections of their PFs, nor is connectivity correlated across separate spatial encounters of a PF with the same PC. In a model of the cerebellar circuit, we then address the possibility that partial connectivity is a substrate of learning, i.e., a fixed, random mask that enforces diversity between PCs. We find that when considering "ensembles" of PCs, random partial connectivity can indeed outperform all-to-all connectivity. Our results provide a theoretical framework for understanding the role of partial connectivity between cerebellar PFs and PCs and may have implications for cerebellum-like systems and beyond.

The brain is the most captivating chef doeuvre of nature. Naturally then, the mind wonders about the process that births such a fascinating organ. Neurodevelopment is a complex yet robust phenomenon that conceals answers to our questions in its intricacies. In an attempt to shed some light on this matter, we study the developing brain connectome of the nematode, C. elegans across the post-embryonic phase. A tiny organism with only around 200 neurons comprising its brain and yet a diverse array of behaviors to display, it makes for a great model. Starting with most of its head neurons already present at hatching, the worm brain accumulates numerous more synaptic connections increasing the edge density. It maintains a weak connectivity throughout thereby, balancing global communication as well as hierarchy. At the mesoscopic level, we find that the core has a conserved backbone of persistent neurons along with a dynamic component formed of transient/recurring neurons. Moreover, the connectome has a rich club organization since the early stage which selectively strengthens indicating progressively denser connectivity among the integrators due to the previously reported asymmetric synapse addition. This asymmetry also shows up in the preservation of input hubs across development and the progressively more centralized organization of the in-degree k-core. Our work provides a new perspective into the neurodevelopment of the brain that may facilitate our understanding of its functioning.

The hippocampal CA2 region is critical for social novelty recognition memory--the discrimination of whether a conspecific is novel or familiar. However, its role in forming a memory of a pair-bonded mate is unknown. To examine how social memories of pair-bonded individuals are encoded, we sought to understand if CA2 and the neighboring CA1 region participate in the memorization and recognition of a pair-bonded mate in monogamous Peromyscus californicus (California mice). Here, we report that CA2 and CA1 show distinct changes in social encoding of an opposite sex conspecific following pair-bonding. Using multi-channel silicon probes, we recorded single units from CA2 and CA1 in freely behaving male mice before and after pair bond formation during interactions with novel and partner females. We found that the strength of CA2 representations of a novel female mouse weakened after pair bond formation, indicating that CA2 may be preferentially important for novelty detection. In contrast, CA1 demonstrated an increase in the strength of encoding a female partner after pair-bond formation, suggesting that CA1 may encode partner memory. These findings indicate that pair bonding shifts the discrimination of social information from CA2 to CA1.

The awake thalamus is known to be able to filter primary sensory input with and without external modulation. Through the construction and analysis of a novel computational model of a lateral geniculate thalamocortical neuron, we demonstrate how the processing of sensory retinal input is influenced by the underlying thalamic dynamic state. Our model, using only currents verified against expression data from publicly available datasets, is the first to produce five experimentally established distinct dynamic firing regimes. We demonstrate that the thalamocortical cell transitions between these dynamic states in response to glutamatergic signals from the cortex or cholinergic arousal signals coming from the brainstem. We focus on signal processing in the model dynamic states associated with the awake thalamic alpha rhythm where we find that the ability of retinal inputs to generate thalamic spikes is a balance between the timing of retinal spikes, the excitability break imposed by the M-current, and the decay time of the L-type calcium current. Finally, we explore how these two currents help the thalamus process extra-retinal rhythmic inputs, showing the model produces entrainment to slower inhibitory and excitatory rhythms, as well as detailing the importance of nesting faster frequency rhythms within slow cycles for successful thalamic transmission. Our results suggest that the awake alpha rhythm is indirectly causal by acting as a marker for the interaction of these two currents. This biophysically-constrained lateral geniculate thalamocortical cell model generates predictions regarding rhythmic dynamics under different arousal states, thalamic control of retinogeniculate transmission, and the possible impacts neurological disorders, like schizophrenia, have on thalamic processing. Variations of this model could be used to explore the functions of higher order thalamic nuclei, thereby extending its use to investigating more complex cognitive processes. Author summaryThe thalamus generates multiple distinct brain rhythms, processes primary sensory inputs, and modulates its output using feedback signals. Previous computational models of the thalamus have typically focused on a subset of these three thalamic functions without drawing relationships among them. Here we present a novel computational thalamic cell model that unites these thalamic processes. We focus on the awake alpha rhythm, a well known thalamic oscillation, and show that it is a signature of a critical working state that enables the experimentally observed thalamic filtering of retinal signals. Additionally, we find this state is optimal for processing and passing non-sensory rhythmic signals. Our model generates testable predictions about which ionic currents control the transmission of external signals. It highlights the roles of two currents from our model that do not have specified functions in the awake thalamus in previous computational models. The work concludes with hypotheses about why neurological disorders that perturb the thalamus from this alpha rhythm working state lead to significant processing errors locally within the thalamus and globally within the brain.

The computational study of epileptic seizure dynamics has primarily focused on the identification of seizure onset zones and propagation pathways. Here, we present a network dynamical model implemented on the empirically measured mesoscale mouse brain network that reveals previously unresolved organizational principles underlying seizure propagation. Rather than the conventional assumption of a single dominant pathway to synchronization, the model reveals multiple competing pathways to synchronization with distinct dynamical properties including transition propensity, recruitment speed, spatial coverage, synchronization stability and transition kinetics. Consequently, node ablation does not uniformly suppress synchronization across pathways, but instead selectively alters pathway occupancy, producing non-trivial alterations to seizure dynamics with potential implications for resection and network-targeted intervention strategies. Biologically, the olfactory and limbic sub-networks emerge as key mesoscale regulators of synchronization dynamics, with olfactory recruitment preferentially constraining global synchronization while limbic-driven pathways preferentially support seizure generalization. More broadly, these findings extend transient explosive synchronization theory by demonstrating that synchronization in biologically constrained networks may emerge through competing mesoscale recruitment programs rather than a single transition process. Together, these findings introduce a new conceptual framework for seizure propagation, suggesting that pathological synchronization emerges not through a single dominant route, but through competing mesoscale dynamical pathways whose accessibility depends on both network architecture and ongoing network state. Significance statementEpileptic seizure propagation is conventionally understood as progressing through a dominant pathway that recruits increasingly larger portions of the brain into pathological synchronization. Using a network dynamical model implemented on the empirical mesoscale mouse connectome, we show that seizure-like synchronization instead emerges through multiple competing pathways with distinct spatial and temporal characteristics. These pathways differ in their propensity for generalization, synchronization stability and sensitivity to node perturbation, such that network interventions selectively reshape pathway accessibility rather than uniformly suppressing seizure dynamics. Our findings introduce a new framework for understanding seizure propagation, identify mesoscale mechanisms linking network architecture to synchronization dynamics and suggest that competing synchronization pathways may represent an important organizing principle in complex brain networks. Graphical abstract O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=97 SRC="FIGDIR/small/730931v1_ufig1.gif" ALT="Figure 1"> View larger version (36K): org.highwire.dtl.DTLVardef@7c4076org.highwire.dtl.DTLVardef@16c0961org.highwire.dtl.DTLVardef@1dbd8a0org.highwire.dtl.DTLVardef@6b06de_HPS_FORMAT_FIGEXP M_FIG C_FIG

The tuft dendrites of layer 5 neurons can support regenerative events - dendritic spikes - that have been proposed to coordinate context-specific engagement and plasticity within cortical networks. However, it remains unclear whether tuft spikes are accompanied by input activity with dynamics that could support these network-level functions. To address this, glutamatergic synapses and postsynaptic calcium signals were simultaneously imaged in the tuft dendrites of layer 5 extratelencephalic neurons within the premotor cortex of mice performing a cued directional licking task. Trial-to-trial, the generation of tuft spikes was associated with a multiphasic elevation in synaptic activity spanning hundreds of milliseconds. This activity was highly specific to the dendrite in which a spike was detected, suggesting the concurrent activation of select subnetworks. Synapses that were strongly coupled to the overall population were the most synchronized with tuft spikes and preferentially encoded the transition between the preparation and action epochs of the task. Even among these strongly coupled synapses, increases in activity were largely specific to synapses located on the spiking dendrite. Surprisingly, among synapses with the poorest population coupling, a second population of coactive synapses was discovered that was also associated with tuft spikes and functionally selective for task-outcome. These results suggest that tuft spikes may be particularly driven by inputs from neurons that are both embedded in sparse subnetworks and synchronized through coupling to larger-scale functional networks. Significance StatementFlexible behavior and learning may depend on interactions between activity in different brain networks and dendritic spikes generated within the neurons that make up the output layer of the neocortex. The results of this study indicate that at the moment of spike generation, spikes in different dendrites are associated with the activation of very specific networks. Yet, across time, the inputs most associated with dendritic spikes are broadly coactive and share selectivity for similar features of behavior. This suggests that dendritic spikes may be particularly driven by the activation of "hub" neurons that coordinate communication between large-scale and small-scale functional networks.