bioRxiv Subject Collection: Neuroscience's Journal

Identifying Neuroimaging Biomarkers of Resilience and Vulnerability to Chronic Stress in An Animal Model: An Exploratory Analysis
Stress is a main contributor to mood disorders, with individuals displaying great heterogeneity in response to stressful life events and adversity. Identifying biomarkers of vulnerability and resilience to stress would facilitate a prevention-based approach to mental illness that benefits individuals and reduces healthcare costs. The present study adopted a multivariate machine-learning approach to track neuroimaging biomarkers predictive of resilience and vulnerability following a chronic restraint stress (CRS) model. 96 male Sprague-Dawley rats underwent two sessions of MRI and behavioral tests, before and after CRS. Resilience and vulnerability to CRS were assessed with elevated plus maze and forced swimming tests. Hierarchical clustering was applied to construct brain networks. Partial correlation was used to compute network connectivity. Repeated nested cross validation with a support vector machine was employed to identify rs-fMRI biomarkers predictive of resilience and vulnerability following CRS. No strong group effect size of behavioral changes following CRS was observed within the same animals, suggesting the presence of resilient and vulnerable subgroups. Although the average model performance was modest (area under the receiver operating characteristic curve: 0.3 ~ 0.67), baseline functional connectivity across cerebellum, brainstem, striatum, prefrontal and salience-orbitofrontal regions, as well as functional alteration across hippocampus, striatum, prefrontal regions, auditory thalamus, cerebellum, inferior colliculi and brainstem were identified as stable features. The present study is the first to identify connectome-based neuroimaging biomarkers predictive of resilience and vulnerability using an animal model. The results may provide insights into neuroimaging biomarkers to aid diagnosis and prevention of mood disorders in humans.

A functional trade-off between executive control and implicit statistical learning is dynamically gated by mind wandering
Human cognition must balance goal-directed behavior with the need to learn from environmental regularities. Mind wandering (MW), a state of attentional decoupling from the task at hand, is paradoxically associated with both executive failures and enhanced implicit statistical learning, yet the direct relationship between these phenomena remains unclear. Here, we provide direct behavioral evidence for a functional trade-off between these competing demands. Using a task that concurrently measured response inhibition, statistical learning, and self-reported task focus, we show that MW is associated with impaired inhibitory control but enhanced learning of probabilistic sequences. Critically, we reveal that these effects are mechanistically linked: the magnitude of the learning enhancement during MW is quantitatively modulated by the efficacy of response inhibition. These findings demonstrate that transient lapses in top-down executive control directly facilitate the implicit extraction of environmental statistics, supporting neurocompetition models and framing MW as a cognitive state that may be evolutionarily preserved to promote the unsupervised acquisition of predictive models.

Music Scaffolds Visual Statistical Sequence Learning Through Network-Level Reorganization in the Brain
Statistical learning - the ability to extract patterns from noisy continuous experiences - is fundamental to human cognition. Yet, how contextual factors shape this process remains poorly understood. Music is an important example of such contextual factors, because it is ubiquitous in human experience and provides a rich temporally-structured stimulus that can co-occur with other learning processes. Here we demonstrate that pairing music fundamentally enhances visual statistical learning, and this is correlated with systematic reorganization of large-scale brain networks. Using fMRI and a novel probabilistic sequence learning paradigm, we show that familiar melodies significantly improved participants' ability to segment continuous visual streams into events and learn sequential relationships. Neuroimaging analyses revealed that the presence of music fundamentally altered the neural network organization that coordinates learning mechanisms: while sequence learning in silence engaged frontal-parietal networks associated with explicit pattern extraction, providing musical temporal structure as a context shifted learning toward MTL-vmPFC circuits recently implicated in schema-guided memory processing. Machine learning analyses confirmed these architectural differences, with the music condition achieving optimal neural prediction of behavioral performance through distributed connectivity patterns while control condition relied on concentrated processing. Our findings support a Cross-Modal Temporal Scaffolding Theory, demonstrating that structured temporal context signals from one modality (here, music) can create more efficient neural states for sequence processing in another through dual mechanisms: enhanced memory integration through schema-guided learning and reduced demands on explicit control resources. These results identify network-level principles for optimizing statistical learning, with broad implications for understanding how environmental context shapes human learning capacity.

Basal forebrain and neural correlates of self-regulation traits in sustained attention
Self-regulation is a human trait consistently associated with success in both academic and professional settings and to better mental health. Based on previous findings, we used functional imaging data in a sustained attention tasks to test three hypotheses on neural substrates associated with individual differences in self-regulation. The first linked higher self-regulation and cognitive control, predicting modulation of recruitment of prefrontal substrates. The second, originating in the animal literature, suggests increased recruitment of cholinergic substrates in the basal forebrain. The third predicted higher modulation of reward-sensitive regions in the brainstem in less regulated individuals for differences in reward levels during the task. The second hypothesis was confirmed by our study, which also provided suggestive evidence for the third hypothesis. Our data suggest that one mechanism of higher self-regulation in man may ensue from greater activity in the cholinergic system to sustain attention during a cognitively simple task.

Testing the auditory steady-state response (ASSR) to 40-Hz and 27-Hz click trains in children with autism spectrum disorder and their first-degree biological relatives: A high-density electroencephalographic (EEG) study
Motivation: Altered auditory processing likely contributes to core social and attentional impairments in autism spectrum disorder (ASD). The auditory steady-state response (ASSR), a neural measure of auditory processing and cortical excitatory-inhibitory balance, has yielded mixed results in ASD. This study uses high density electroencephalography (EEG) to evaluate ASSR in ASD and unaffected siblings to clarify neural mechanisms underlying auditory deficits in autism. Methods: High-density 70-channel EEG was recorded in children (8-12 years, IQ >80) with ASD (n=53), typically developing (TD) peers (n=35), and unaffected biological siblings (n=26) during 500-ms binaural click trains (27- and 40-Hz) in an active oddball task. Results: No group differences were observed in frequency-following responses (FFR) to 27- or 40-Hz stimuli, although higher 40-Hz power was associated with older age and better behavioral performance in ASD. The broad-band response from 180-250 ms was reduced in ASD for both stimulation frequencies, particularly in the low-frequency (<8 Hz) range, and significantly correlated with IQ and age. Siblings showed intermediate broad-band responses. Discussion: While FFRs appeared intact in ASD, we observed reduced broad-band response in the transition period to the steady state FFR, which was specific to low (<8-Hz) frequencies, potentially reflecting reduced synchronization at timescales that correspond with slower, syllabic rhythms (~4-8 Hz) occurring in natural speech. Intermediate responses in first-degree relatives suggest that this is related to genetic vulnerability for ASD and highlights its clinical relevance. These findings suggest intact sensory processing in ASD alongside possible top-down auditory feedback deficits, which may serve as heritable neurophysiological markers.

Hippocampal-guided reconstruction of an event's prior temporal context
Events are thought to be encoded into memory in the context of temporally-adjacent events. Here, using fMRI, we show that when a visual stimulus from the past is re-encountered, lateral occipitotemporal cortex (LOTC) reinstates visual content from stimuli that were adjacent to the original encounter. This LOTC reinstatement effect was selective to stimuli that were subjectively remembered and was mediated by stimulus-specific activity patterns within the hippocampus.

Geometric principles determining the morphology of oligodendrocyte precursor cells in brain white matter
We used transgenic mice expressing membrane-tagged green fluorescence protein in the oligodendrocyte precursor cells (OPCs), high-resolution imaging, and detailed quantitative morphometric analysis to investigate the geometrical principles that govern structural organization of OPCs in the mouse corpus callosum. Our major findings are: (1) During the first two months of postnatal life in mice, total length of all OPCs processes increases via elaboration of new branches from the existing processes rather than via the appearance of new processes; (2) New branches are preferentially added to more distal sites of OPCs processes; (3) The processes of OPCs show stronger preferential alignment with the posterior-anterior brain axis rather than with the lateral-medial or dorsal-ventral brain axes; at the same time, the processes of OPCs show stronger preferential alignment with the lateral-medial than with the dorsal-ventral brain axis. Our study is the first detailed comprehensive analysis of OPCs morphology comparable to those available for neurons. It helps understanding the geometrical principles that govern structural organization of OPCs. These principles are important when taking into account that OPCs receive synaptic input from neurons and are capable of synaptic integration. Arborization and structural organization of OPCs processes is expected to influence the travel of synaptic input from the processes (where synapses are located) to the cell soma (where synaptic inputs are integrated), in analogy to how it occurs in neurons. Hence, the integrated synaptic signal at the OPCs cell soma which is likely to influence development and behavior of OPCs will depend on the cell morphology.

Motor cortical areas facilitate schema-mediated integration of new motor information into memory
New information is rapidly learned when it is compatible with pre-existing knowledge, i.e. with a previously acquired schematic representation of the learned information. The influence of pre-established schema on learning has been extensively studied in the declarative memory domain, where it was shown that schema-compatible information could be rapidly assimilated into neocortical storage, bypassing the slow hippocampo-neocortical memory transfer process. Schema-mediated learning was recently examined in the motor memory domain; however, its neural substrates remain unknown. The goal of this study was to address this knowledge gap using both univariate and multivariate analyses of functional Magnetic Resonance Imaging (fMRI) data acquired in 60 young healthy participants during the practice of a motor sequence that was either compatible or incompatible with a previously acquired cognitive-motor schema. Consistent with previous literature, our behavioural results suggest that performance of sequential movements was enhanced when practice occurred in a context that was compatible with the previously acquired schema. Brain imaging results show that practice in a schema-compatible context specifically recruited the left primary motor cortex and resulted in a decrease in connectivity between the bilateral motor cortex and a set of task-relevant brain regions including the hippocampus, striatum, and cerebellum. Temporally fine-grained MRI analyses revealed that multivoxel activation patterns in the primary motor and the premotor cortices were modulated by schema-compatibility, with greater pattern similarity detected for sequence elements corresponding to and surrounding novel sequential movements under schema-compatible compared with -incompatible conditions. Altogether, these results suggest that motor cortical regions facilitate schema-mediated integration of novel movements into memory.

Early life chronic stress-disrupted activity of the dorsal raphe nucleus selectively drives behavioral impairments
Stress elicits variable systemic and neural changes in vertebrates, with outcomes ranging from adaptive to pathological. Several studies have implicated the dorsal raphe nucleus (DRN), a brainstem nucleus containing a heterogeneous population of serotonergic (5-HT) neurons, in the adaptive stress response and the pathological changes resulting from chronic stress. However, it is not known whether early life chronic stress affects the developing DRN activity, or whether the stress-induced changes affect 5-HT DRN neurons in a subregion- or phenotype-specific manner. To answer these questions, we used in vivo 2-photon calcium imaging of 5-HT DRN neurons in larval zebrafish exposed to chronic unpredictable stress during early life. We found that early life chronic stress prevented the normal habituation of the serotonergic system to a repeated acute stressor by altering the balance of excitatory/inhibitory responses within the DRN. Interestingly, these changes were most pronounced in a subset of stress-vulnerable serotonergic cells co-expressing GABAergic markers. Further, using chemogenetic ablation of 5-HT DRN neurons, we showed that stress-induced plasticity of the DRN contributed to changes in startle response habituation and in locomotive activity, but not in anxiety-like behaviors. Collectively, our results emphasize the role of stress-induced plasticity of DRN neurons in the selective regulation of maladaptive behavioral outcomes.

Sensitivity analysis of voltage-gated ion channel models.
Modeling voltage-gated ion channel function is essential for understanding neuronal excitability. However, finding the right balance between model complexity and practicality is a significant challenge. In this study, I explored how sensitive Markov models of ion channels were to different parameters, beginning with a straightforward two-state system and progressing to more intricate three- and four-state models. Through tests using step voltage and sine wave protocols, I discovered clear sensitivity patterns during both the activation and deactivation phases. As expected, forward transitions were most influential during activation, while reverse transitions became more pronounced during deactivation. Notably, I found that transitions between closed states had minimal impact on the overall whole-cell current or the opening probability (Po). This underscores a significant limitation in relying on these measurements to constrain specific parameters. These results highlight the strengths and limitations of Markov models, showing that their effectiveness can be restricted by how sensitive parameters are to the protocols used in experiments. My findings set clear guidelines on the level of complexity feasible in Markov models of voltage-gated ion channels and offer insights that can help create more efficient and reliable models of neuronal excitability.

MotilA - A Python pipeline for the analysis of microglial fine process motility in 3D time-lapse multiphoton microscopy data
MotilA is a Python-based image analysis pipeline for quantifying fine process motility of microglia from 3D time-lapse two-channel fluorescence microscopy data. Developed for high-resolution multiphoton in vivo imaging datasets, MotilA enables both single-file and batch processing across multiple experimental conditions. It performs image preprocessing, segmentation, and motility quantification over time, using a pixel-based change detection strategy that yields biologically interpretable metrics such as the turnover rate (TOR) of microglial fine processes. While originally designed for microglial imaging, the pipeline can be extended to other cell types and imaging applications that require analysis of dynamic morphological changes. MotilA is openly available, platform-independent, and includes extensive documentation, tutorials, and example data to facilitate adoption by the broader scientific community. It is released under the GPL-3.0 open-source license.

Walking Selectively Modulates Behaviorally Relevant Auditory Responses According to Stride Cycle Phase
Human perception operates in dynamic environments, requiring sensory processing to continuously adapt to ongoing motor activity. Here, we investigated how natural walking modulates auditory change detection by recording EEG while participants performed a transitional click train task during sitting and walking. Behaviorally, walking significantly elevated auditory change detection thresholds, reflecting reduced perceptual sensitivity. Electrophysiologically, three distinct auditory responses onset, change, and offset were evoked; both the onset and change responses were attenuated during walking, with the most pronounced effect observed for the cognitively demanding change response. Notably, this modulation was further amplified for the change response during active engagement compared to passive listening, implicating attentional resource allocation. Strikingly, both behavioral performance and change-related neural responses fluctuated rhythmically with the gait cycle, while onset and offset responses remained comparatively stable. Together, these findings reveal a stride-phase-dependent gating mechanism that selectively modulates perceptually relevant auditory signals, illustrating how the brain rhythmically reallocates sensory resources to optimize perception during locomotion.

Ethanol induces neuroimmune dysregulation and soluble TREM2 generation in a human iPSC neuron, astrocyte, microglia triculture model
Alcohol use disorders (AUDs) affect substantial populations worldwide and increase the risk of developing cognitive impairments and alcohol-associated dementia. While chronic inflammatory signaling likely plays an important role in alcohol-associated neurological sequalae, the precise mechanisms underlying alcohol-associated neuropathology remain enigmatic. We hypothesize that alcohol leads to neuroimmune dysregulation among neurons, astrocytes, and microglia; and is perpetuated by innate immune signaling pathways involving cell-cell signaling. To investigate how alcohol dysregulates neuroimmune interactions in a human context, we constructed a triculture model comprising neurons, astrocytes, and microglia derived from human induced pluripotent stem cells. After exposure to ethanol, we observed significant differential gene expression relating to innate immune pathways, inflammation, and microglial activation. Microglial activation was confirmed with morphological analysis and expression of CD68, a lysosomal-associated membrane protein and marker for phagocytic microglial activation. A striking finding in our study was the elevation of TREM2 expression and, specifically, TREM2 alternative splice variants that are predicted to give rise to soluble TREM2 . These results suggest that ethanol exposure in the brain may lead to increased microglial activation and production of soluble isoform named TREM2219 through alternate splicing. Deciphering the molecular and cellular mechanisms underpinning ethanol-related neuroimmune dysregulation within a human context promises to shed light on the etiology of AUD-related disorders, potentially contributing to the development of effective therapeutic strategies.

Cellf-Deception: Human microglia clone 3 (HMC3) cells exhibit more astrocyte-like than microglia-like gene expression
Recent advances in Alzheimer's research suggest that the brain's immune system plays a critical role in the development and progression of this devastating disease. Microglial cells are vital as immune cells in the brain's defense system. Human Microglia Clone 3 (HMC3) is a cell line developed as a promising experimental model to understand the role of microglial cells in human diseases, including Alzheimer's and other neurodegenerative diseases. The frequency of HMC3 cell usage has increased in recent years, with the idea that this cell line could serve as a convenient model for human microglial cell functions. Here, we utilize gene-pair ratios from pseudo-bulk and scRNAseq expression data to create predictive models of cell-type origins. Our model reveals that the HMC3 cell line represents various cells, with the highest cell similarity score relating to astrocytes, not microglia. These findings suggest that the HMC3 cell line is not a reliable human microglia model and that extreme caution should be taken when interpreting the results of studies using the HMC3 cell line.

Sardine: a modular framework for developing data acquisition and near real-time analysis applications
We present Sardine, a software framework built with .NET to control experimental setups through the reliable execution of dynamic networks of independent modules, where each module can interface with a hardware device (e.g., camera, motor) or represent an operation over data (e.g., image filter, data stream). The Sardine framework delivers robust fault-tolerant hardware control and eliminates downstream delays by implementing dedicated data processing queues for each module. Any .NET class can be seamlessly adapted into a Sardine module, enabling effortless integration with existing codebases. Sardines modular architecture ensures flexibility to accommodate changes in experimental paradigms, simplifying the adaptation of essential features, such as swapping hardware components or redefining the logic of their interactions. The core of Sardine is a novel aggregation system that connects modules in two layers, designed to streamline complex workflows such as those in microscopy applications. The first layer (link layer) enables concurrent operations between modules, such as synchronizing a camera module with a stage controller or a laser module, while maintaining a dependency tree to ensure devices operate as intended. For example, in an imaging experiment, Sardine oversees individual module malfunctions, such as a camera failure or stage misalignment, gracefully handling errors and dynamically restoring functionality to impacted network segments to minimize disruption. The second layer (data layer) facilitates the transmission of information across modules by associating them with operations that produce, consume, or transform data. In an imaging context, this could involve processing raw image data from a camera module, passing it through a real-time analysis module for feature detection, and forwarding the results to a visualization module for immediate feedback. This two-layer architecture ensures seamless data flow and robust error handling, making Sardine ideal for dynamic and complex experimental setups. Sardine also integrates logging, a metadata collection system, and tools to create graphical applications.

Three-dimensional ultrastructural differences between thalamic and non-thalamic recipient layers in macaque V1
Understanding the synaptic characteristics of each cortical layer is essential for elucidating the functional architecture of each brain region. In the current study, we made a detailed quantitative comparison of the synaptic structure in the predominantly input layers of primate primary visual cortex (layer 4C) and in the predominant output layer (layer 3B) using focused ion beam scanning electron microscopy (FIB/SEM). We quantified the synaptic density in each layer, classified synaptic boutons according to their number of synapses and mitochondrial content, and quantified key morphometric parameters, including bouton volume, postsynaptic density (PSD) area and morphology, volume occupied by mitochondria, and postsynaptic targets. Our results revealed that for all the layers there is a higher proportion of single-synapse boutons without mitochondria. Multisynaptic boutons containing mitochondria (MSBm+) -which likely correspond to TC terminals -were significantly more abundant in the thalamocortical recipient layers 4C and 4C{beta}. These MSBm+ boutons were also larger, more likely to contact dendritic spines, and contained more mitochondria than other bouton categories. In contrast, layer 3B, displayed a lower prevalence of MSBm+ boutons, these boutons were smaller than those in layer 4C and made fewer synapses. These findings highlight laminar differences in bouton architecture and support the idea that TC synapses are structurally adapted to support high synaptic efficacy. Together, our data provide a detailed quantitative framework for understanding the synaptic organization of primate V1, with implications for sensory processing and cortical circuit function.

Test-Retest Reliability Analysis of Resting-state EEG Measures and Their Association with Long-Term Memory in Children and Adults
EEG resting-state measures, such as spectral power and microstates, have been associated with human long-term memory (LTM) performance. However, findings across studies are inconsistent and sometimes contradictory, likely due to a low reliability of the measures employed. These inconsistencies limit the interpretability and generalizability of results, emphasizing the need for a systematic evaluation of measure reliability. In this study, we addressed this gap by identifying the most reliable EEG resting-state measures and evaluating their predictive value for LTM performance in a second-language (L2) vocabulary learning paradigm. A group of children (N = 36) and adults (N = 90) participated in 2 studies on app-assisted learning of second language vocabulary. Participants completed a test on L2 vocabulary and a resting-state EEG recording (180 s eyes open) before and after learning a second language using a smartphone app. We used Intraclass Correlation Coefficients (ICC) to identify resting state EEG measures with satisfying test-retest reliability (ICC >= 0.75) and then assessed how these reliable measures are associated with L2 vocabulary learning representing LTM performance. Highest ICC values were found for oscillatory power in the alpha range and in the frequency of occurrences, duration and coverages of microstates. Calculations yielded ICC values of 0.84/0.86 (children/adults) for alpha power and 0.88/0.80 for microstate measures. Of these measures, only alpha power showed a positive correlation with LTM performance, but only in the adult population (r = 0.38, p < .01). No other measures were associated with LTM (all p > 0.05). Alpha power could thus serve as a stable and reliable marker of the neural mechanisms accounting for high LTM performance in the fully developed adult brain.

Temporal Dynamics of Depression in Premanifest Huntington's Disease: A Network Dysconnection Approach
Depression emerges years before motor signs in Huntington's disease gene expansion carriers (HDGECs), yet the neural mechanisms underlying this temporal relationship remain unclear. A growing body of evidence links depression in HDGECs to striatal and default mode network (DMN) regions, however, the longitudinal evolution of network dysfunction and its relationship to depressive symptoms has not been characterized. The study included 105 HDGECs (53 Females, Mage = 43, Mean CAG repeat length = 43) from the largest longitudinal study of premanifest HD, TrackOn-HD. Here, we applied spectral dynamic causal modelling to examine effective connectivity changes over 24 months, alongside voxel-based morphometry of grey matter atrophy and linear mixed models of depressive symptoms, including volumetry. We show that depression-related network dysconnection operates independently of regional grey matter atrophy, with HDGECs exhibiting widespread striatal volume loss but no differential atrophy patterns between depression groups. Critically, we demonstrate that larger posterior cingulate cortex volumes predicted increased depression severity specifically in HDGECs with depression history, across both Beck Depression Inventory-II (p = .041) and Hospital Anxiety and Depression Scale-Depression Subscale (p = .002). Spectral dynamic causal models explained 88.6% of variance in depression groups and 90.4% in non-depression groups, demonstrating robust model performance. Longitudinal effective connectivity analyses revealed distinct dysconnection profiles: HDGECs with depression history showed widespread interhemispheric alterations including progressive inhibitory changes of striatal-DMN circuits and aberrant hippocampal regulatory control, while those without depression exhibited more focal dysconnection. These patterns occurred despite no group differences in grey matter atrophy trajectories across striatal or cortical regions over the 24-month period. Furthermore, we identify that clinically elevated depressive symptoms associate with differential connectivity patterns of the PCC and hippocampus depending on depression history, suggesting pathological mechanisms become progressively maladaptive. Our findings challenge that structural preservation confers functional resilience in neurodegeneration, instead suggesting pathological mechanisms and revealing that functional network reorganization in relatively preserved regions drives depression vulnerability in premanifest HD.

Miniaturized widefield microscope for high speed voltage imaging
Functional imaging in freely moving animals with genetically encoded voltage indicators (GEVIs) will open new capabilities for neuroscientists to study the behavioral relevance of neural activity with high spatial and temporal precision. However, miniaturization of an imaging system with sufficient collection efficiency to resolve the small changes in fluorescence yield from voltage spikes, as well as development of efficient image sensors that are sufficiently fast to capture them, has proven challenging. We present a miniaturized microscope designed for voltage imaging, with a numerical aperture of 0.6, 250 m field of view and 1.3 mm working distance that weighs 16.4 g. We show it is capable of imaging in vivo voltage spikes from Voltron2 with a spike peak-to-noise ratio >3 at a framerate of 530 Hz.

A Novel Infrared-Based Non-Invasive Brain Stimulation Approach for Neurological Disorders
Non-invasive neuromodulation techniques are gaining attention as alternatives to traditional deep brain stimulation for treating neurological disorders. This study explores the use of short-wave infrared light to stimulate deep-brain regions without surgical intervention. By leveraging advanced photon modeling and optical simulations, the proposed system optimizes light penetration for effective neural activation. Experimental evaluations demonstrate its feasibility in delivering targeted stimulation while maintaining safety and efficiency. These findings highlight the potential of infrared-based neuromodulation for future clinical applications in neurotherapy and cognitive enhancement.

Primary cortical neurons form calcium-phosphate sheets with a bone-precursor-like ultrastructure
Calcium-phosphate is a critical component in healthy bone and teeth formation, but its pathologic buildup in brain can occur in dyshomeostatic calcium disorders like Alzheimer's disease and Leigh syndrome. Extracellular calcium-phosphate in the nervous system is not well understood, but prior evidence suggests mitochondria as a potential source. Recently, large unidentified double-membrane-bound hyper-electron dense sheet aggregates of unknown content were reported in Huntington's disease model neurons using cryo-ET that were absent in wild-type control neurons. We use a combination of cryo-ET, cryo-CLEM, and LDSAED to demonstrate that these sheet aggregates are generated by wild-type cortical neurons isolated from embryonic day 18 rat embryos and have an octacalcium phosphate-like structure. They appear to be derived from mitochondria and are extruded at least in part by migrasomes. These findings reveal an important link for how mitochondria can serve as reservoirs for intracellular and extracellular calcium-phosphate highlighting them as potential therapeutic targets in neurological disorders characterized by pathological calcification, such as Alzheimer's disease.