bioRxiv Subject Collection: Neuroscience's Journal

Learning Residual-based Biomarkers of Cognitive Health via Self-Supervised Learning on EEG State Transitions
Deep learning (DL) models have achieved impressive performance in EEG-based prediction tasks, but they often lack interpretability, limiting their clinical utility. In this study, we introduce a novel self-supervised learning (SSL) framework inspired by neurophysiological reactivity. Our approach models healthy EEG transitions between ocular states by predicting an EEG-derived feature under eyes-open conditions, using features from eyes-closed recordings. The residual between observed and predicted values quantifies deviations from normative brain dynamics and serves as a candidate biomarker. To improve clinical relevance, we propose two optimisation strategies that promote residuals predictive of pathology. We evaluated the framework using healthy cohorts (LEMON, Dortmund Vital) for self-supervised training, and the AI-Mind cohort for downstream prediction of plasma p-tau217 levels, a proxy for cognitive pathology. Despite extensive hyperparameter optimisation, predictive performance remained poor across all methods, including baseline models, suggesting limitations in the downstream proxy or input signal. Nonetheless, our approach provides a transparent methodology for transition-based EEG biomarker discovery grounded in self-supervised learning.

Domain-specific functions of LRIT3 in synaptic assembly and retinal signal transmission
LRIT3 is a leucine-rich repeat (LRR) protein that is expressed in the retina, and its absence causes complete congenital stationary night blindness (cCSNB), a genetically diverse disorder characterized by impaired low-light vision, myopia, and nystagmus. LRIT3 is expressed in rod and cone photoreceptors, and it trans-synaptically organizes the assembly of the glutamate signaling complex, the signalplex, on depolarizing bipolar cells (DBCs). LRIT3 is a single-pass membrane protein with extracellular LRR, IG, and FN3 domains. The mechanism by which LRIT3 controls postsynaptic receptor organization remains unknown. We address this by using rAAV to express deletion constructs in LRIT3 knockout retinas and examining LRIT3 trafficking, as well as the structural and functional recovery of the signalplex in DBCs. We show the LRR domain is required for trafficking LRIT3 to the synapse in cones, but not rod photoreceptors, although it is needed for reassembly and function of the rod BC signalplex. Neither the IG nor the FN3 domain is needed for synaptic localization of LRIT3. However, the IG domain is required for the localization of TRPM1 to the signalplex and thus function. The FN3 domain is not necessary for either DBC signalplex assembly or function. Our data demonstrates that the LRR and IG domains of LRIT3 are crucial for TRPM1 localization and retinal function, with the LRR domain playing a key role in the differential function of LRIT3 at rod and cone synapses. Notably, our results show that restoring Nyctalopin localization to the DBC signalplex alone is insufficient to restore TRPM1 expression. Based on our findings, we propose a model in which the LRR domain trans-synaptically binds with Nyctalopin, while the IG domain interacts with TRPM1.

Dream content and slow waves benefit preys against predators in a video game confrontation
Every studied animal species exhibits some form of sleep, a state evolved under the sustained influence of prey and predator behaviors. Dreams have been proposed to have evolved as offline mental simulations able to warn against impending threats from the environment, such as predators. To assess how the roles of prey and predator interact with sleep and dreaming, we used a first-person shooter game as a proxy for predator-prey confrontations. Electroencephalographic (EEG) and electrocardiographic (ECG) signals were recorded from 30 human adults paired and simultaneously recorded while 1) playing a video game round against each other, in which one participant was armed with a gun and the other was not, 2) taking a nap, 3) reporting dreams and/or thoughts, and 4) playing another round with the same gun assignment. We found that the participants in the prey role were highly affected by sleep and dreaming. Their score gains were positively correlated with slow wave activity during the nap, with how much the dreams were related to the game, and with heart rate variability during the first round. In contrast, no significant correlations were found for participants in the predator role. The results suggest that slow wave activity and game-related dream content during sleep improve the post-sleep performance of individuals under stressful, prey-like situations.

Unifying spatial and episodic representations in the hippocampus through flexible memory use
A key question in neuroscience is why the hippocampus is essential for episodic memory in humans, while dominantly exhibiting spatial representations in a number of other species. Some accounts suggest that spatial representation is the primary hippocampal function. Here, we propose that the primary function is storing and retrieving episodic memories, and spatial representations emerge due to this memory function. To demonstrate this, we adopt a computational model that autonomously learns to store information in memory and retrieve it to solve a variety of tasks. In memory tasks, the model develops associations and categorical representations akin to concept cells. In navigation tasks, the model forms representations of the spatial structure, performs geometric computations, and even learns representations of unique events similar to recently discovered barcodes. Our model predicts that the hippocampus represents any task-relevant variable, if the animal learns the task, suggesting that space is not special for the hippocampus.

Chronic delivery of buprenorphine during abstinence decreases incubation of heroin seeking and neuronal activation in medial prefrontal cortex and striatum in male and female rats
Rationale and Objective: Buprenorphine is an FDA-approved medication for opioid addiction, but the brain regions underlying its therapeutic effects remain unknown. We previously found that chronic buprenorphine treatment decreases several relapse-related behaviors in rats. Here, we tested whether chronic buprenorphine decreases the time-dependent increase (incubation) in heroin seeking during abstinence. We also used the activity marker Fos to test whether buprenorphine inhibition of incubation of heroin seeking is associated with decreased activation of cortical and striatal regions. Methods: We trained Oprm1-Cre rats and their wild-type littermates to self-administer heroin for 12 days (6 h/day). On abstinence Day 1, we tested for heroin seeking under extinction conditions. On Day 14, we implanted osmotic minipumps containing saline or buprenorphine (6 mg/kg/day). On abstinence Days 21-22, we tested the rats (or not tested, baseline condition) for incubation of heroin seeking, after which brains were collected for Fos immunohistochemistry. Results: Oprm1-Cre rats did not differ from wild-type littermates in heroin self-administration or non-incubated heroin seeking on abstinence Day 1. In both genotypes, chronic buprenorphine decreased incubated heroin seeking on Days 21-22. Buprenorphine also decreased incubation-related neuronal activation in several cortical areas (anterior cingulate and dorsal peduncular, but not prelimbic, infralimbic, or orbitofrontal cortex) and striatal areas (dorsolateral and dorsomedial striosomes and nucleus accumbens core, but not shell). Conclusions: Chronic buprenorphine decreased incubation of heroin seeking, supporting the predictive validity of the rat model. This effect was associated with decreased neuronal activity in specific subregions of the medial prefrontal cortex and striatum.

Hierarchical and Spatial Mapping of Whole-Brain c-Fos Activity Reveals Distinct Opioid and Withdrawal Neuronal Ensembles
How opioid exposure and withdrawal states shape brain activity at the systems and circuit level remains poorly understood. Here, we use whole-brain, cellular-resolution c-Fos mapping to define brain-wide activity patterns and neuronal ensembles associated with morphine administration and withdrawal. To account for the brain's anatomically nested structure, we developed and applied a hierarchical statistical framework that detects region-specific changes in activity and outperforms conventional methods that treat brain regions as individual, unrelated units. These distributed signals formed ensembles with consistent and anatomically structured patterns of activity, both within subregions and across multiple connected brain areas. By combining TRAP2-based activity tagging with acute whole-brain c-Fos staining, we identified morphine- and withdrawal-activated ensembles and found that they are largely non-overlapping at the single-cell level, even within the same brain region. Integration with existing spatial transcriptomics datasets identified molecular markers for these state-specific ensembles in key brain areas such as the nucleus of accumbens, amygdala and ventral tegmental areas. Lastly, by integrating Allen mouse whole-brain transcriptional datasets, we identified the molecular identity of the morphine- administration and withdrawal ensembles. These findings define dissociable neuronal ensembles that encode opposing drug states and introduce a scalable framework for linking whole-brain activity to molecular and circuit-level mechanisms.

PEDIA-BRAIN: A single nuclei multiomic encyclopedia of the human pons provides a resource for normal development and disease vulnerability
The human pons relays sensory and motor information between the brain and the body. It is a key site for pathological processes, including diffuse midline gliomas (DMGs) and multiple sclerosis (MS) which predominantly arise in childhood and middle age, respectively. Although multiple studies address disease states, a comprehensive resource for normal pons development is lacking. Here we present PEDIA-BRAIN, an encyclopedia of gene expression and chromatin accessibility from 149,771 human pons nuclei spanning the first trimester to early adulthood to serve as a resource for the scientific community. Exploration of the encylopedia identified two trajectories to mature oligodendrocytes and developmental restriction of synaptic interactions between neurons and oligodendrocyte precursor cells. To illustrate the utility of the resource, we mapped single cell transcriptomes from DMG and MS samples and identified depletion of mature oligodendrocyte subtypes as a shared feature of both diseases. Data may be browsed and downloaded at pediabrain.nchgenomics.org.

The CaMKII D135N mutation blocks kinase activity and reduces GluN2B binding
Three recent studies claimed that induction of long-term potentiation (LTP) of synaptic strength requires structural rather than enzymatic functions of the Ca2+/Calmodulin(CaM)-dependent protein kinase II (CaMKII). One study utilized the CaMKII D135N mutation, which was claimed to abolish enzymatic activity without affecting the structural function, i.e. binding to GluN2B. We found here that the D135N mutant indeed abolished enzymatic kinase activity and autophosphorylation at T286 (pT286). However, D135N mutation additionally reduced binding to GluN2B and prevented persistence of co-condensation with GluN2B. Similar as with the T286A mutant, GluN2B binding of the D135N mutant could be partially rescued by AS283, an inhibitor that directly enhances GluN2B binding. This partial effect on GluN2B binding has to be taken into account when using D135N to distinguish between enzymatic versus structural functions of CaMKII. Nonetheless, as discussed here, the D135N mutant indeed supports a structural rather than enzymatic CaMKII function in LTP induction.

The link between gender inequality and the distribution of brain regions' relative sizes across the lifespan and the world
Evidence is emerging that the socioeconomic environment in general, and gender inequality in particular, can be a shaping force on brain structure. However, our understanding of the nature of this influence throughout the lifespan is often limited because most current data sets are geographically and demographically narrow, making it unclear whether results hold across distinct world populations. Here we analyse, for the first time, data from an online MRI analysis platform comprising 13277 subjects from 52 countries and the five continents, across ages that range from childhood to late life. We examined how gender inequality, jointly examined with economic inequality, relates to differences in brain grey matter between males and females. We found that the association between female-male brain differences and gender inequality increases with age, suggesting a cumulative effect of gender inequality throughout life. Further, by considering additional variables that are specifically related to the economy, we found that this effect was, as per current data, dominated by the economic aspects of inequality.

The Highly Selective 5-HT2B Receptor Antagonist MW073 Mitigates Aggressive Behavior in an Alzheimer Disease Mouse Model.
Alzheimer disease (AD) is a multifactorial neurodegenerative disorder and the leading cause of dementia worldwide. Progressive synaptic and neuronal loss underlies the decline in cognition and daily functioning, often accompanied by behavioral and psychological symptoms. Among these, neuropsychiatric disturbances such as agitation and aggression affect 20%-65% of patients and represent a major source of caregiver burden. The serotonin receptor antagonist MW073 has recently emerged as a potential therapeutic candidate, showing efficacy in counteracting Abeta- and tau-induced synaptic and memory deficits in AD mouse models. Here, we investigated whether MW073 also mitigates aggressive behavior in Tg2576 mice, a widely used AD model that also displays heightened aggressiveness. Our findings demonstrate that MW073 significantly reduced aggressive tendencies in Tg2576 mice, suggesting that serotonergic modulation may represent a promising strategy to address both cognitive and neuropsychiatric symptoms of AD.

Photosensitivity of an aging brain
Red light is considered less phototoxic than blue light and is widely used in both research and photobiomodulation therapy. The difference in the response to brain exposure to light between young and old mammals is currently unknown. We found that brain exposure to blue light caused local damage in the cerebral cortex in both young and old mice. Brain exposure to red light did not have any noticeable effect on young mice. However, it caused a marked reduction in electroencephalogram power, damaged fiber bundles throughout the brain, and brought about a coma-like state in old mice. The effect of red light on electroencephalogram power was dose-dependent and particularly strong in the theta range. When delivered at a lower intensity but over a longer period, red light produced a similar reduction in electroencephalogram power and brain damage as those seen in the mice treated with higher irradiation over a shorter period. These results indicate that the impact of light on electroencephalogram and brain tissue strongly depends not only on the light wavelength, duration and intensity of the exposure, but also on the age of the animal and type of tissue exposed to light.

Sparkling Water Limits Cognitive Fatigue with Stable Pupil Diameter during Real World Esports Training
Prolonged esports play induces cognitive fatigue with reduced executive function and pupil constriction as a neurobiomarker of prefrontal activity across expertise levels. Since this pupil-linked cognitive fatigue is mitigated by sparkling water intake in casual players, we hypothesized that sparkling water prevents cognitive fatigue and sustains esports performance in esports athletes under real training conditions that reflect competitive practice. To test this hypothesis, nineteen hardcore esports players participated in a randomized crossover trial. Each completed two 3-hour sessions of head-to-head virtual football matches while consuming either sparkling water or plain water. Subjective fatigue, enjoyment, and executive function (flanker task) were assessed, and pupil diameter, heart rate, and salivary cortisol were continuously monitored. In-game performance was also recorded. Accuracy declined and pupil constriction progressed under the plain water condition, whereas executive accuracy was preserved and pupil diameter was maintained over the three-hour session in the sparkling water condition. Subjective ratings of fatigue and enjoyment increased similarly in both conditions. Heart rate, including both mean and peak values, and salivary cortisol concentrations showed no significant differences between conditions. In terms of in-game performance, goals, passes, and interceptions did not differ between conditions, whereas the number of fouls was lower and the number of shots was higher in the sparkling water condition. These findings demonstrate that sparkling water mitigates cognitive fatigue and sustains esports performance in esports athletes under real training conditions. As a caffeine- and sugar-free beverage, sparkling water may serve as a practical, low-risk intervention for managing cognitive fatigue and supporting daily practice in modern esports settings.

Beta-burst dynamics in the motor cortex are reshaped through sensorimotor refinement
Beta-band activity (15-30 Hz) in motor cortex is closely linked to movement-related processing, yet its transient burst dynamics during long-term learning remain poorly understood. We used High-density electroencephalography (HD-EEG) to examine how beta bursts change over nine sessions of a bimanual coordination task, under adaptive and non-adaptive training conditions. Both training conditions led to motor skill learning, with the adaptive group improving more during training, but the non-adaptive group showing better retention. With training, contralateral primary motor cortex showed stronger beta desynchronization during movement and greater synchronization after movement. These changes reflected underlying burst dynamics: post-movement bursts became more temporally confined and consistent, with increased probability and reduced timing variability across sessions. Only the adaptive group showed a session-related increase in burst amplitude. These results demonstrate that beta burst features reorganize with practice, providing a temporally precise neural readout of training progression and revealing how different learning conditions shape cortical dynamics over time.

Auditory midbrain encodes training-induced plasticity in soundlocalization behavior
The ability of the brain to learn from experience and to compensate for changes in sensory inputs is usually associated with plasticity in the cerebral cortex. Descending corticofugal pathways have been implicated in learning but there is limited evidence that subcortical processing is shaped by training. We show that sound-source location can be accurately decoded from neuronal populations recorded in the inferior colliculus of one or both hemispheres while ferrets perform a localization task. Furthermore, changes in neural decoding performance matched improvements in localization accuracy of individual animals when ferrets were trained to adapt to abnormal spatial cues resulting from reversible occlusion of one ear. These findings demonstrate that the activity patterns of populations of neurons in the inferior colliculus can account for sound localization behavior in different hearing contexts, and that training-dependent plasticity in the auditory midbrain may support spatial learning following monaural hearing loss.

Sensorimotor remapping drives task specialization in prefrontal cortex
The prefrontal cortex (PFC) plays a key role in flexible, context-dependent decision-making. Yet, the population-level mechanisms that support this flexibility remain unclear: is there a generalist neural population involved in representing useful information across tasks, or are representations distributed over multiple modules of task-specialized subpopulations? Here, we investigate whether the population structure modularity in medial prefrontal cortex (mPFC) increases with sensorimotor remapping between task rules in two different paradigms. In a paradigm where all stimuli were remapped, we found a highly modular structure with two task-specialized subpopulations and one generalist. In a different paradigm where only half of the stimuli were remapped, modularity decreased but remained significant. Based on predictions drawn from a recurrent neural network fitted to experimental data, we propose that a gain modulation by global inhibition of mPFC can explain how task-specialized neurons emerge to varying degrees in each paradigm.

Neonatal Microglia and Their Secretome as Mediators of Brain Repair
Microglia are essential regulators of myelin integrity and repair, yet their regenerative capacity declines with ageing and in neurodegenerative diseases such as multiple sclerosis (MS). Neonatal microglia retain a uniquely reparative program that may offer insight into restoring lost functions in the adult CNS. Here we show that transplantation of neonatal microglia ameliorates disability, reduces leukocyte infiltration, and promotes remyelination in both inflammatory (EAE) and non-inflammatory (cuprizone) models, and reverses cognitive decline in aged mice. These benefits persisted even when transplanted cells remained confined to the meninges and were reproduced by the neonatal microglia secretome, indicating a paracrine mechanism. Multi-omic profiling revealed that the neonatal secretome is enriched in trophic factors and membrane-building lipids compared to adult microglia, while transcriptomic analyses of treated aged brains showed reactivation of developmental repair pathways and suppression of inflammatory signatures. Together, these results demonstrate that neonatal microglia re-engage rejuvenation-like programs in the adult CNS and highlight the importance of multifactorial strategies, integrating trophic, metabolic, and immunomodulatory cues, over single-target approaches. Our findings establish early microglial programs as a paradigm for designing new regenerative therapies for CNS disorders.

Brain-like variability in convolutional neural networks reveals evidence-,uncertainty- and bias-driven decision-making
Even when stimuli and tasks are held constant, brain activity fluctuates markedly across trials, yet it is not well understood how these fluctuations affect decision-making and behavior. Here we address this gap in knowledge using a convolutional neural network (CNN) trained on a perceptual decision-making task. Applying an established analytical framework, we show that CNN activity exhibits trial-level variability reflecting findings from human neuroimaging. Examination of the internal state revealed three distinct activation patterns reflecting: (i) decisions dominated by strong sensory evidence, (ii) decisions under low discriminability with weak or ambiguous evidence, and (iii) decisions in which sensory evidence was opposed by bias. Notably, choice bias shifted the decision boundary, improving performance despite low or conflicting evidence. Together, these findings show that variability in CNN activity can serve as a model for understanding how the brain transforms changes in internal states into adaptive behavior, providing a bridge between fluctuations in activity, decision-making, and behavior.

Complementary attentional mechanisms for the resolution of representational ambiguity in the human brain
When we view cluttered environments containing multiple objects, the neural code for individual items can become ambiguous. This reflects a capacity limit in spatially-tuned visual neurons: when multiple objects fall within the cell receptive field, output cannot be attributed to a single object. Vision models from primate electrophysiology propose this is resolved by attention, which biases competition between object representations. However, evidence in humans is sparse and inconclusive, and reliance on single-cell data limits insight into underlying mechanisms and neurophysiological scope. Here, we use a novel multivariate approach to the analysis of human EEG and concurrent EEG/MRI to address this. First, we test whether attention is recruited by representational ambiguity. Second, we identify the mechanisms that act on representations of attended and unattended objects to resolve ambiguity. Finally, we characterize the millisecond timing and whole-brain action of these mechanisms to identify pervasive effects in semantic and executive brain networks.

Memory dysfunction and psychiatric outcomes in anti-NMDA receptor encephalitis are linked to altered structural brain complexity
Introduction: Most patients with anti-N-methyl-D-aspartate receptor encephalitis (NMDARE) experience long-term neuropsychiatric sequelae despite immunotherapy. However, how these residual symptoms relate to structural brain changes in the post-acute phase remains unclear. Recently, fractal dimensionality (FD) has emerged as a sensitive imaging marker of structural brain complexity in related conditions but has not been explored in autoimmune encephalitis. Methods: This cross-sectional study combined clinical, cognitive, and neuroimaging analyses in 70 patients with post-acute NMDARE (median time from onset: 22 months), and 70 healthy controls matched for age (t=-0.57, p=0.57) and sex ({chi}2=0, p=1). High-resolution T1-weighted magnetic resonance imaging (MRI) data were analyzed with FreeSurfer and computational fractal analysis. Clinical outcomes were assessed with the modified Rankin Scale (mRS) and Clinical Assessment Scale in Autoimmune Encephalitis (CASE). Psychiatric manifestations underwent phenotypical analyses, and memory performance was assessed with standardized neuropsychological tests. Results: Patients with NMDARE were severely affected at peak illness but showed substantial overall improvement in the post-acute stage (median mRS at peak=5, post-acute=1, z=-7.2, p<0.001; median CASE at peak=11, post-acute=1, z=-7.2, p<0.001). However, 67% of patients showed residual CASE symptoms, with memory dysfunction (61%) and psychiatric symptoms (36%) representing the most prevalent domains. Therein, psychiatric symptoms showed a phenotypical shift from a 'schizophrenia-like' phenotype at peak illness to an affective phenotype in the post-acute stage. Neuroimaging uncovered a characteristic pattern of reduced structural brain complexity, including the hippocampus bilaterally and a fronto-cingulo-temporal cluster in cortical gray matter and cerebral white matter. Importantly, normative modeling revealed that patients with residual symptoms showed stronger alterations of brain complexity than those without (psychiatric: t=-2.65, p=0.010; memory: t=-3.98, p<0.001; both vs. no symptoms: t=-5.46, p<0.001). Similarly, patients with stronger alterations of brain complexity showed lower scores of visuospatial and verbal memory (all pFDR<0.015). Discussion: Post-acute NMDARE is characterized by systematic reductions in structural brain complexity, consistently involving previously implicated regions while identifying changes in the cingulate cortex as a new morphological correlate. These changes are linked to residual psychiatric symptoms and memory dysfunction, highlighting FD as a promising new imaging marker of long-term outcomes. Our findings suggest that current treatment strategies may be insufficient to fully address the residual symptom burden after the acute phase of NMDARE.

Evaluating the quality of brainstem ROI registration using structural and diffusion MRI
Accurate registration of regions of interest (ROIs) from standard atlases to participants' native spaces is a critical step in fMRI studies, as it directly affects the reliability of sampled BOLD signals. While T1-weighted (T1w) image-based ROI registration is well validated and widely adopted in cortical fMRI, its performance degrades in brainstem studies due to the small size, dense packing, and poor visibility of brainstem nuclei on T1w contrast. We hypothesized that incorporating diffusion MR images, containing more information about internal brainstem architecture, should improve ROI registration accuracy. To test this, we developed four registration pipelines that either included or excluded diffusion-based alignment components and evaluated their performance using data from n=10 healthy participants. Registration accuracy was assessed using Dice coefficients for the red nucleus (RN) and mis-registration fractions for the dorsal raphe nucleus (DRN). The results showed that diffusion-based pipelines, using fractional anisotropy (FA) images, non-diffusion-weighted (b0) images, and multivariate combination, outperformed the T1w-only baseline. Probabilistic maps derived from inverse-transformed native ROIs further supported improved sensitivity to inter-individual anatomical variability in the diffusion-augmented pipelines. In addition, analysis of gradient magnitude maps from the Jacobian determinants revealed associations between localized deformation and image modality-specific landmarks. These findings demonstrate the potential of diffusion-augmented pipelines for improving brainstem ROI registration, which could enhance the robustness of fMRI studies on brainstem disorders characterized by functional dysregulation.

VMAT2 dysfunction impairs vesicular dopamine uptake, driving its oxidation and α-synuclein pathology in DJ-1-linked Parkinson's disease neurons
Parkinson's disease (PD) is characterized by -synuclein accumulation and dopaminergic neuron degeneration, with dopamine (DA) oxidation emerging as a key pathological driver. However, the mechanisms underlying this neurotoxic process remain unclear. Using PD patient-derived and CRISPR-engineered iPSC midbrain dopaminergic neurons lacking DJ-1, we identified defective sequestration of cytosolic DA into synaptic vesicles, which culminated in DA oxidation and -synuclein accumulation. In-depth proteomics, state-of-the-art imaging, and ultrasensitive DA probes uncovered that decreased VMAT2 protein and function impaired vesicular DA uptake, resulting in reduced vesicle availability and abnormal vesicle morphology. Furthermore, VMAT2 activity and vesicle endocytosis are processes dependent on ATP, which is notably reduced in DJ-1-deficient dopaminergic neurons. ATP supplementation restored vesicular function and alleviated DA-related pathologies in mutant dopaminergic neurons. This study reveals an ATP-sensitive mechanism that regulates DA homeostasis through VMAT2 and vesicle dynamics in midbrain dopaminergic neurons, highlighting enhanced DA sequestration as a promising therapeutic strategy for PD.

Growth inhibitory effect of cerebrospinal fluid of multiple sclerosis patients on HFF2 cells
Multiple sclerosis (MS) is an autoimmune disease characterized by demyelination of the central nervous system (CNS). Occurrence of immunogenic cell death (ICD) in this disease and expression of ICD markers in the CSF of MS patients are documented. However, cytotoxic effect of MS patients cerebrospinal fluid (CSF) on the cells is not reported. To evaluate the cytotoxic impact of CSF from MS patients on cell viability, human foreskin fibroblast (HFF2) cells were treated with CSF samples from individuals with MS and healthy controls. The results inferred that patient-derived CSF significantly declined the survival of HFF2 cells in comparison to the CSF of healthy individuals. This observation implies a cytotoxic environment and characteristic of patients CSF, which could be due to the presence of cellular-damaging factors such as calreticulin (CRT) in the patients CSF. The cytotoxic environment of CSF might reflect ICD occurrence in CNS of the patients and also could explain the progressive essence of the disease.

Molecular Deconstruction of the Medullary Raphe Magnus in the rat: Transcriptional Responses to Repeated Seizures
Sudden Unexpected Death in Epilepsy (SUDEP) is a leading cause of death in patients with epilepsy and is thought to result from dysfunctional and/or failure of cardiorespiratory control systems. Post-mortem brainstem tissue analyses in human SUDEP cases point to reductions in markers of the brainstem serotonin (5-HT) system, which is known as a brainstem center that provides excitatory neuromodulation. We have previously shown in a knockout rat model (SSkcnj16-/- rats) that repeated seizures led to progressively greater ventilatory inhibition in the post-seizure period, seizure-associated mortality, and reduced brainstem 5-HT and tryptophan hydroxylase (Tph) particularly within the Raphe Magnus (RMg). Here, we account for the cellular constituency and local transcriptional responses to repeated seizures in male SSkcnj16-/- rats that experienced daily seizures for 3, 5, 7, or 10 consecutive days using single nuclear RNA sequencing (snRNA seq) from brainstem tissue biopsies including the RMg (-12.12 mm to -10.30 mm caudal to Bregma) two hours post-seizure. Unbiased cluster analysis identified 18 cell major clusters that were by identified by the expression of known gene markers, with most cells being oligodendrocytes. However, local RMg neurons showed the greatest numbers of differentially expressed genes with seizures compared to all other cell types. Further re-clustering of neuronal cell types yielded 14 distinct RMg neuron subpopulations, including 5 types of GABAergic neurons, 2 glutamatergic clusters, and 2 groups of 5-HT neurons which all had unique expression profiles. Nearly all DEGs across neuronal subtypes were increased following seizures, and a large fraction of which were common across seizure days and across neuron type suggesting uniformity in cellular response to seizures in this region. These studies provide foundational information regarding the cellular constituency of the RMg region in the rat, and altered neuronal function following repeated seizures in the absence of changes in other cell types in this key region of cardiorespiratory control.

Temporal Interference Stimulation Can Enhance or Disrupt Human Memory Encoding as a Function of Brain Location and Frequency
Visual memory relies on synchronized interactions and rhythms between the medial temporal lobes and neocortical brain regions. Non-invasive manipulation of memory-related brain regions, specifically deeper temporal lobe regions, has been limited by the lack of precision of non-invasive neuromodulation; when targeting deeper structures, the cortex is always stimulated, never deeper structures in isolation. Temporal Interference (TI) stimulation, a novel non-invasive technique, uses high-frequency carrier fields to deliver targeted, physiologically relevant neuromodulation via amplitude-modulated envelopes at specific brain regions. Here, we investigate TI s impact on figure memory encoding in 70 healthy participants using the Rey-Osterrieth and Taylor Complex Figure tasks, with TI applied in several brain regions independently and simultaneously; allowing investigation of combinations of medial temporal lobe and neocortical brain regions. Interestingly, higher frequency TI envelopes (130 Hz offset) targeting bilateral hippocampi and temporal cortices significantly impair recall (p = 6.54e-04), while lower frequency TI envelopes (5 Hz offset) targeting only the bilateral hippocampi significantly enhance recall (p = 0.0447). Stimulation using other combinations of medial temporal lobe and neocortical regions showed no effect, underscoring the critical role of frequency and focality of non-invasive brain stimulation and correct target selection. Finally, functional MRI reveals strong differences between the effects of 130 Hz and 5 Hz envelopes, specifically in hippocampal BOLD signals, brain connectivity, default mode, and attentional networks. These findings demonstrate TI s ability to bidirectionally modulate memory encoding through precise frequency and target tuning, offering a powerful tool for cognitive neuroscience and potential therapeutic applications for memory disorders.

Dynamic and context-dependent modulation of proprioceptive input in primate primary motor cortex
During voluntary movement, somatosensory input is attenuated -sensory gating - which may prevent the CNS from being overwhelmed by predictable afferent feedback not essential for motor control. In the cerebral cortex, sensory gating has been demonstrated in primary (S1) and higher-order sensory areas. The primary motor cortex (M1) is the major cortical output relay to the spinal cord and muscles, and proprioceptive feedback is crucial for shaping motor output, much as in S1. However, whether proprioceptive signals to M1 are attenuated during movement, and if so, why, remains unclear. We recorded somatosensory-evoked potentials (SEPs) in M1 of two monkeys performing a wrist task while electrically stimulating muscle and cutaneous afferents innervating forearm extensors and adjacent skin. Both local field potentials and single-neuron recordings revealed significant suppression of muscle-evoked SEPs during Active movement and Static hold phases, providing direct evidence that proprioceptive input to M1 is generally gated during motor execution, as previously reported for cutaneous input. Yet, amidst this suppression, SEPs in a subset of M1 neurons were preserved during Static hold, especially those evoked by antagonist muscle afferents. Because monkeys had to maintain precise joint angles for stable posture, these results suggest that proprioceptive signals from antagonist muscles - likely reflecting spindle activity in lengthening muscles - escape attenuation to provide information essential for joint angle control. Overall, our findings demonstrate that while proprioceptive input to M1 is broadly suppressed during motor output, specific afferent signals from antagonist muscles are selectively maintained in a context-dependent manner to support posture control.

Single nucleus multiomic atlas of human dorsal root ganglia reveals the contribution of non-neuronal cell types to pain
Sensory neurons residing in dorsal root ganglia (DRG) transmit sensory information such as pain, itch, touch, pressure and bodily position to the central nervous system. The activity of sensory neurons is regulated by non-neuronal cell types in the DRG, including satellite glial cells (SGCs) and fibroblasts. Dysregulated gene expression in DRG cells contributes to sensory nervous system disorders such as chronic pain and itch. Understanding the genetic underpinnings of these conditions requires dissecting transcriptional regulation in human DRG (hDRG). In this study, we profiled transcriptomic and chromatin accessibility landscapes from postmortem hDRG samples at single-cell level. We demonstrate that sequencing depth significantly impacts downstream analysis, with deeper sequencing yielding more detected cells and features, improved data integration, refined clustering and annotation, and more accurate scientific interpretations. We identified nine major cell types, defined their molecular signatures, and mapped cis-regulatory landscapes. Integration of gene expression with chromatin accessibility enabled peak-gene association and transcriptional network analyses, revealing transcription factors, their target genes, and their regulatory elements. This approach uncovered cell types, genes, and cis-regulatory regions potentially driving pain conditions. Our unbiased genome-wide analysis confirmed known pain-related genes and highlighted novel candidates. These findings provide new insight into molecular mechanisms and candidate cell types involved in pain. Importantly, our results demonstrate that non-neuronal cell types, including endothelial cells, fibroblasts, macrophages, and SGCs, play critical roles in pain pathogenesis and should be investigated as therapeutic targets.

Transmitted alpha-synuclein extracellular vesicles downregulate axonal flux of retrograde carriers in recipient neurons
Alpha-synuclein (alpha-syn) is a cytosolic protein located in nerve terminals and is involved in several neurodegenerative diseases such as Parkinson's disease. Recent studies have demonstrated that alpha-syn can be transmitted from neuron to neuron via exosomal release, thereby contributing to the propagation of alpha-syn pathology. However, the mechanism by which alpha-syn-containing exosomes perturb the function of recipient neurons is currently unknown. Retrograde axonal transport of carriers emanating from the presynapse is essential for neuronal survival. To determine the effect of transmitted alpha-syn on neuronal retrograde trafficking, we used conditioned medium from alpha-syn transfected donor hippocampal neurons, applied to naive (non-transfected) recipient neurons cultured in microfluidic chambers. Time-lapse imaging of retrograde carriers containing cholera toxin B subunit (CTB) was then performed in these recipient neurons. Here, we show that conditioned medium from alpha-synWT-GFP transfected donor neurons significantly downregulated the frequency of retrograde CTB carriers when applied to recipient neurons. This effect was abolished by (i) inhibiting endocytosis in recipient neurons, (ii) blocking exosomal release from donor cells via an Hsp90-dependent mechanism, or (iii) by using conditioned medium from neurons transfected with a Parkinson's mutant (alpha-synA30P-GFP). Whilst transmitted, alpha-synWT-mEos2, alpha-synA30P-mEos2, and mEos2 alone were detected in recipient neurons using single-molecule imaging. Interestingly, alpha-synWT-mEos2 exhibited lower mobility and periodic (190 nm) immobilizations along the axon. Our data suggest that the downregulation of vesicular trafficking by transmitted alpha-syn does not rely on the specific release nor uptake of exosomal alpha-syn but probably depends on its interaction with endogenous alpha-syn following endocytosis in recipient neurons. The transmitted alpha-syn-containing extracellular vesicles therefore control essential axonal trafficking in recipient neurons, an effect lost with the Parkinson's disease mutant alpha-synA30P.

PhaseICA: Complex-Valued Decomposition of Spatially Independent Brain Waves
Understanding complex dynamics from spatiotemporal signals requires robust tools capable of decoding reoccurring patterns. Traditional ICA methods often overlook the spatially non-stationarity nature of brain activity across both the frequency and spatial domains. We propose a novel data-driven approach, named PhaseICA, designed to extract reoccurring spatiotemporal patterns, referred to as brain waves. Unlike conventional ICA methods that focus solely on amplitude, PhaseICA incorporates phase information directly into the component estimation, preserving the nonstationary property that real-valued ICA methods typically discard. The method performs spatial independence optimization in the complex domain by minimizing a complex entropy bound over the eigenvectors of Hilbert-transformed signals. The proposed method captures spatial propagation across brain regions with interpretable and compact representations, offering a promising foundation for decoding brain dynamic systems and revealing the temporal relationship of regions.

Internal mechanosensory feedback modulates central pattern generation to coordinate ovulation in Drosophila
Successful reproduction requires precise coordination of muscle contractions during ovulation and has been well investigated in insects. However, the neural mechanisms of the central pattern generator (CPG) orchestrating this process remain poorly understood. Using Drosophila, we identify a novel pair of multidendritic mechanosensory neurons (mdn-LO) in the lateral oviducts that express the mechanosensitive channels TMC (transmembrane channel-like protein), PPK (pickpocket), and Piezo. We demonstrate that TMC is essential for ovulation coordination: tmc mutation or neuron-specific knockdown significantly increases egg-jamming at the junction between the lateral and the common oviduct, while ppk or Piezo single mutants reduce egg-laying without causing jamming. Calcium imaging reveals that mdn-LO neurons are rhythmically activated by oviduct contractions, and chemogenetic stimulation of these neurons triggers muscle contraction. Trans-Tango and GRASP (GFP Reconstitution Across Synaptic Partners) analyses show that mdn-LO neurons form synapses with efferent insulin-like peptide 7 (ILP7)-expressing motor neurons in the abdominal ganglion. Activating either mdn-LO or ILP7 neurons induces egg-jamming, indicating their roles within a CPG network. This circuit, including the TMC-expressing mdn-LO neurons and a set of ILP7 neurons, organizes an innate behavior by coupling strategic sensory information and a CPG to ensure bilateral coordination of oviduct contractions to prevent egg jamming during oviposition.

Effects of the Neutral CB1 Receptor Antagonist AM6527 on Spontaneous, Consummatory, and Motivated Behavior in Mice
Rationale. The cannabinoid type-1 receptor (CB1R) signaling pathway plays a central role in regulating motivational and feeding behaviors. Neutral CB1R antagonists represent a promising therapeutic class with potentially fewer adverse effects than inverse agonists, yet their behavioral effects remain incompletely characterized. Objectives. We investigated the behavioral profile of AM6527, orally bioavailable neutral CB1R antagonist, across naturalistic and operant paradigms in male mice. To evaluate dopaminergic involvement in AM6527s effects, we employed several pharmacological interventions. Results. Using machine learning-based Motion-Sequencing (MoSeq), which parses spontaneous behavior into sub-second syllables, we found that AM6527 did not affect overall speed in an open field, however, it increased the self-directed behaviors and reduced specific locomotor syllables at the highest dose tested. In a naturalistic reward consumption paradigm, AM6527 produced a dose-dependent reduction in milk intake. Operant conditioning paradigms revealed robust suppression of motivated responding on fixed ratio-3 and progressive ratio (PR) schedules for palatable milk reward, with the greatest impact on high-baseline performers under PR conditions. To understand the dopaminergic involvement, we co-administered dopaminergic drugs (targeting D1R, D2R, or dopamine transporter) which resulted in partial rescue of operant responding, indicating dopaminergic and non-dopaminergic contributions to AM6527s observed behavioral effects. Conclusion. Our findings suggest that neutral CB1R antagonism suppresses consummatory and motivated behaviors via dopamine-dependent and -independent mechanisms. By leveraging sub-second behavioral analysis with MoSeq, we further reveal distinct changes in spontaneous behavior, underscoring the relevance of CB-based treatments for maladaptive appetitive and motivational states in both psychiatric and metabolic disorder.